CFA Level 2 2023 Summaries

Quantitative Methods
Learning Module 1 Basics of Mul ple Regression and Underlying Assump ons

• Mul ple linear regression is used to model the linear rela onship between one dependent variable
and two or more independent variables.

• In prac ce, mul ple regressions are used to explain rela onships between nancial variables, to test
exis ng theories, or to make forecasts.

• The regression process covers several decisions the analyst must make, such as iden fying the
dependent and independent variables, selec ng the appropriate regression model, tes ng if the
assump ons behind linear regression are sa s ed, examining goodness of t, and making needed
adjustments.

• A mul ple regression model is represented by the following equa on: Yi =b0 +b1X1i +b2X2i +b3X3i
+...+bkXki +εi,i=1,2,3,...,n, where Y is the dependent variable, Xs are the independent variables from 1
to k, and the model is es mated using n observa ons.

• Coe cient b0 is the model’s “intercept,” represen ng the expected value of Y if all independent
variables are zero.

• Parameters b1 to bk are the slope coe cients (or par al regression coe cients) for independent
variables X1 to Xk. Slope coe cient bj describes the impact of independent variable Xj on Y, holding
all the other independent variables constant.

• There are ve main assump ons underlying mul ple regression models that must be sa s ed,
including (1) linearity, (2) homoskedas city, (3) independence of errors, (4) normality, and (5)
independence of independent variables.

• Diagnos c plots can help detect whether these assump ons are sa s ed. Sca erplots of dependent
versus and independent variables are useful for detec ng non-linear rela onships, while residual
plots are useful for detect- ing viola ons of homoskedas city and independence of errors.
ti
ffi
ti
ti
ti
ti
ti
fi
ti
ti
ti
ffi
ti
ti
ffi
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ffi
ti
ti
fi
fi
tt
ti
ti
ti
fi
 Learning Module 2 Evalua ng Regression Model Fit and Interpre ng Model Results

• In mul ple regression, adjusted R2 is used as a measure of model goodness of t since it does not
automa cally increase as independent variables are added to the model. Rather, it adjusts for the
degrees of freedom by incorpora ng the number of independent variables.

• Adjusted R2 will increase (decrease) if a variable is added to the model that has a coe cient with an
absolute value of its t-sta s c greater (less) than 1.0.

• Akaike’s informa on criterion (AIC) and Schwarz’s Bayesian informa on criteria (BIC) are also used to
evaluate model t and select the “best” model among a group with the same dependent variable. AIC
is preferred if the purpose is predic on, BIC is preferred if goodness of t is the goal, and lower values
of both measures are be er.

• Hypothesis tests of a single coe cient in a mul ple regression, using t-tests, are iden cal to those in
simple regression.

• The joint F-test is used to jointly test a subset of variables in a mul ple regression, where the
“restricted” model is based on a narrower set of independent variables nested in the broader
“unrestricted” model. The null hypothesis is that the slope coe cients of all independent variables
outside the restricted model are zero.

• The general linear F-test is an extension of the joint F-test, where the null hypothesis is that the slope
coe cients on all independent variables in the unrestricted model are equal to zero.

• Predic ng the value of the dependent variable using an es mated mul ple regression model is
similar to that in simple regression. First, sum, for each independent variable, the es mated slope
coe cient mul plied by the assumed value of that variable, and then add the es mated intercept
coe cient.

• In mul ple regression, the con dence interval around the forecasted value of the dependent variable
re ects both model error and sampling error (from forecas ng the independent variables); the larger
the sampling error, the larger is the standard error of the forecast of Y and the wider is the con dence
interval.
fl
ffi
ffi
ffi
ti
ti
ti
ti
ti
fi
ti
tt
ti
ti
ti
fi
ffi
ti
ti
ti
ti
ti
ffi
ti
ti
ti
ti
fi
fi
ti
ti
ti
ffi
fi
 Learning Module 3 Model Misspeci ca on

• Principles for proper regression model speci ca on include economic reasoning behind variable
choices, parsimony, good out-of-sample performance, appropriate model func onal form, and no
viola ons of regression assump ons.

• Failures in regression func onal form are typically due to omi ed variables, inappropriate form of
variables, inappropriate variable scaling, and inappropriate data pooling; these may lead to the
viola ons of regression assump ons.

• Heteroskedas city occurs when the variance of regression errors di ers across observa ons.
Uncondi onal heteroskedas city is when the error variance is not correlated with the independent
variables, whereas condi onal heteroskedas city exists when the error variance is correlated with the
values of the independent variables.

• Uncondi onal heteroskedas city creates no major problems for sta s cal inference, but condi onal
heteroskedas city is problema c because it results in underes ma on of the regression coe cients’
standard errors, so t-sta s cs are in ated and Type I errors are more likely.

• Condi onal heteroskedas city can be detected using the Breusch–Pagan (BP) test, and the bias it
creates in the regression model can be corrected by compu ng robust standard errors.

• Serial correla on (or autocorrela on) occurs when regression errors are correlated across
observa ons and may be a serious problem in me-series regressions. Serial correla on can lead to
inconsistent coe cient es mates, and it underes mates standard errors, so t-sta s cs are in ated
(as with condi onal heteroskedas city).

• The Breusch–Godfrey (BG) test is a robust method for detec ng serial correla on. The BG test uses
residuals from the original regression as the dependent variable run against ini al regressors plus
lagged residuals, and H0 is the coe cients of the lagged residuals are zero.

• The biased es mates of standard errors caused by serial correla on can be corrected using robust
standard errors, which also correct for condi onal heteroskedas city.

• Mul collinearity occurs with high pairwise correla ons between independent variables or if three or
more independent variables form approximate linear combina ons that are highly correlated.
Mul collinearity results in in ated standard errors and reduced t-sta s cs.

• The variance in a on factor (VIF) is a measure for quan fying mul collinearity. If VIFj is 1 for Xj, then
there is no correla on between Xj and the other regressors. VIFj > 5 warrants further inves ga on,
and VIFj > 10 indicates serious mul collinearity requiring correc on.

• Solu ons to mul collinearity include dropping one or more of the regression variables, using a
di erent proxy for one of the variables, or increasing the sample size.
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fl
ffi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fl
ti
ti
ti
ti
ti
ffi
ti
fl
fi
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
tt
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ffi
fl
ti
ti
 Learning Module 4 Extensions of Mul ple Regression

• Two kinds of observa ons may poten ally in uence regression results: (1) a high-leverage point, an
observa on with an extreme value of an independent variable, and (2) an outlier, an observa on with
an extreme value of the dependent variable.

• A measure for iden fying a high-leverage point is leverage. If leverage is k+1 greater than 3(k+1/n),
where k is the number of independent variables, then the observa on is poten ally in uen al. A
measure for iden fying an outlier is studen zed residuals. If the studen zed residual is greater than
the cri cal value of the t-sta s c with n – k – 2 degrees of freedom, then the observa on is
poten ally in uen al.

• Cook’s distance, or Cook’s D (Di), is a metric for iden fying in uen al data points. It measures how
much the es mated values of the regression change if observa on i is deleted. If Di > 2√k/n, then it is
highly likely to be in uen al. An in uence plot visually presents leverage, studen zed residuals, and
Cook’s D for each observa on.

• Dummy, or indicator, variables represent qualita ve independent variables and take a value of 1 (for
true) or 0 (for false) to indicate whether a speci c condi on applies, such as whether a company
belongs to a certain industry sector. To capture n possible categories, the model must include n – 1
dummy variables.

• An intercept dummy adds to or reduces the original intercept if a speci c condi on is met. When the
intercept dummy is 1, the regression line shi s up or down parallel to the base regression line.

• A slope dummy allows for a changing slope if a speci c condi on is met. When the slope dummy is 1,
the slope changes to (dj + bj) × Xj, where dj is the coe cient on the dummy variable and bj is the
slope of Xj in the original regression line.

• A logis c regression model is one with a qualita ve (i.e., categorical) dependent variable, so logis c
regression is o en used in binary classi ca on problems, which are common in machine learning and
neural networks.

• To es mate a logis c regression, the logis c transforma on of the event probability (P) into the log
odds, ln[P/(1 − P)], is applied, which linearizes the rela on between the transformed dependent
variable and the independent variables.

• Logis c regression coe cients are typically es mated using the maximum likelihood es ma on (MLE)
method, and slope coe cients are interpreted as the change in the log odds that the event happens
per unit change in the independent variable, holding all other independent variables constant.
ti
ti
ti
ti
ti
ti
ti
fl
ft
ti
ti
ti
ti
fl
ti
ffi
ffi
ti
ti
ti
ti
fl
ti
ti
fi
ti
ti
ti
ft
fl
ti
fi
ti
ti
ti
fi
ffi
ti
ti
ti
fl
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
fl
ti
ti
ti
ti
ti
 Learning Module 5 Time-Series Analysis

• The predicted trend value of a me series in period t is bˆ0 + bˆ1t in a linear trend model; the
predicted trend value of a me series in a log-linear trend model is e bˆ0+bˆ1t.

• Time series that tend to grow by a constant amount from period to period should be modeled by
linear trend models, whereas me series that tend to grow at a constant rate should be modeled by
log-linear trend models.

• Trend models o en do not completely capture the behavior of a me series, as indicated by serial
correla on of the error term. If the Durbin–Watson sta s c from a trend model di ers signi cantly
from 2, indica ng serial correla on, we need to build a di erent kind of model.

• An autoregressive model of order p, denoted AR(p), uses p lags of a me series to predict its current
value: xt = b0 + b1xt−1 + b2xt−2 + . . . + bpxt–p + εt.

• A me series is covariance sta onary if the following three condi ons are sa s ed: First, the expected
value of the me series must be constant and nite in all periods. Second, the variance of the me
series must be constant and nite in all periods. Third, the covariance of the me series with itself for
a xed number of periods in the past or future must be constant and nite in all periods. Inspec on
of a nonsta onary me-series plot may reveal an upward or downward trend (non constant mean)
and/or nonconstant variance. The use of linear regression to es mate an autoregressive me-series
model is not valid unless the me series is covariance sta onary.

• For a speci c autoregressive model to be a good t to the data, the autocorrela ons of the error term
should be 0 at all lags.

• A me series is mean rever ng if it tends to fall when its level is above its long-run mean and rise
when its level is below its long-run mean. If a me series is covariance sta onary, then it will be mean
rever ng.

• The one-period-ahead forecast of a variable xt from an AR(1) model made in period t for period t + 1
is xˆt+1 = bˆ0 + bˆ1 xt. This forecast can be used to create the two-period-ahead forecast from the
model made in period t, xˆt+2 = bˆ0 + bˆ1 xt+1. Similar results hold for AR(p) models.

• In-sample forecasts are the in-sample predicted values from the es mated me-series model. Out-of-
sample forecasts are the forecasts made from the es mated me-series model for a me period
di erent from the one for which the model was es mated. Out-of-sample forecasts are usually more
valuable in evalua ng the forecas ng performance of a me-series model than are in-sample
forecasts. The root mean squared error (RMSE), de ned as the square root of the average squared
forecast error, is a criterion for comparing the forecast accuracy of di erent me-series models; a
smaller RMSE implies greater forecast accuracy.

• Just as in regression models, the coe cients in me-series models are o en unstable across di erent
sample periods. In selec ng a sample period for es ma ng a me-series model, we should seek to
assure ourselves that the me series was sta onary in the sample period.
fi
ff
ti
ti
ti
ti
fi
ti
ti
ti
ft
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ffi
ti
ti
fi
ti
fi
ti
ti
fi
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ff
fi
ft
ti
ti
ti
ti
ti
fi
ti
ff
ti
ti
fi
ti
ff
ti
 • A random walk is a me series in which the value of the series in one period is the value of the series
in the previous period plus an unpredictable random error. If the me series is a random walk, it is
not covariance sta onary. A random walk with dri is a random walk with a nonzero intercept term.
All random walks have unit roots. If a me series has a unit root, then it will not be covariance
sta onary.

• If a me series has a unit root, we can some mes transform the me series into one that is
covariance sta onary by rst-di erencing the me series; we may then be able to es mate an
autoregressive model for the rst-di erenced series.

• An n-period moving average of the current and past (n − 1) values of a me series, xt, is calculated as
[xt + xt−1 + . . . + xt−(n−1)]/n.

• A moving-average model of order q, denoted MA(q), uses q lags of a random error term to predict its
current value.

• The order q of a moving-average model can be determined using the fact that if a me series is a
moving-average me series of order q, its rst q autocorrela ons are nonzero while autocorrela ons
beyond the rst q are zero.

• The autocorrela ons of most autoregressive me series start large and decline gradually, whereas the
autocorrela ons of an MA(q) me series suddenly drop to 0 a er the rst q autocorrela ons. This
helps in dis nguishing between autoregressive and moving-average me series.

• If the error term of a me-series model shows signi cant serial correla on at seasonal lags, the me
series has signi cant seasonality. This seasonality can o en be modeled by including a seasonal lag in
the model, such as adding a term lagged four quarters to an AR(1) model on quarterly observa ons.

• The forecast made in me t for me t + 1 using a quarterly AR(1) model with a seasonal lag would be
xt+1 = bˆ0 + bˆ1 xt + bˆ2 xt−3.

• ARMA models have several limita ons: The parameters in ARMA models can be very unstable;
determining the AR and MA order of the model can be di cult; and even with their addi onal
complexity, ARMA models may not forecast well.

• The variance of the error in a me-series model some mes depends on the variance of previous
errors, represen ng autoregressive condi onal heteroskedas city (ARCH). Analysts can test for rst-
order ARCH in a me-series model by regressing the squared residual on the squared residual from
the previous period. If the coe cient on the squared residual is sta s cally signi cant, the me-
series model has ARCH(1) errors.

• If a me-series model has ARCH(1) errors, then the variance of the errors in period t + 1 can be
predicted in period t using the formula
ti
ti
ti
ti
ti
fi
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
fi
fi
ti
ti
ffi
ff
ti
ti
ff
ti
ti
fi
ti
ti
ti
ft
fi
ti
ft
ffi
ti
ti
ft
ti
ti
ti
ti
ti
fi
ti
ti
fi
ti
ti
ti
ti
ti
ti
fi
ti
ti
 • If linear regression is used to model the rela onship between two me series, a test should be
performed to determine whether either me series has a unit root:
If neither of the me series has a unit root, then we can safely use linear regression.
If one of the two me series has a unit root, then we should not use linear regression.
If both me series have a unit root and the me series are cointegrated, we may safely use linear
regression; however, if they are not cointegrated, we should not use linear regression. The (Engle–
Granger) Dickey–Fuller test can be used to determine whether me series are cointegrated.
ti
ti
ti
ti
ti
ti
ti
ti
 Learning Module 6 Machine Learning

Machine learning methods are gaining usage at many stages in the investment management value chain.
Among the major points made are the following:

• Machine learning aims at extrac ng knowledge from large amounts of data by learning from known
examples to determine an underlying structure in the data. The emphasis is on genera ng structure
or predic ons without human interven on. An elementary way to think of ML algorithms is to “ nd
the pa ern, apply the pa ern.”

• Supervised learning depends on having labeled training data as well as matched sets of observed
inputs (X’s, or features) and the associated out- put (Y, or target). Supervised learning can be divided
into two categories: regression and classi ca on. If the target variable to be predicted is con nuous,
then the task is one of regression. If the target variable is categorical or ordinal (e.g., determining a
rm’s ra ng), then it is a classi ca on problem.

• With unsupervised learning, algorithms are trained with no labeled data, so they must infer rela ons
between features, summarize them, or present underlying structure in their distribu ons that has not
been explicitly provided. Two important types of problems well suited to unsupervised ML are
dimension reduc on and clustering.

• In deep learning, sophis cated algorithms address complex tasks (e.g., image classi ca on, natural
language processing). Deep learning is based on neural networks, highly exible ML algorithms for
solving a variety of supervised and unsupervised tasks characterized by large datasets, non-lineari es,
and interac ons among features. In reinforcement learning, a computer learns from interac ng with
itself or data generated by the same algorithm.

• Generaliza on describes the degree to which an ML model retains its explanatory power when
predic ng out-of-sample. Over ng, a primary reason for lack of generaliza on, is the tendency of
ML algorithms to tailor models to the training data at the expense of generaliza on to new data
points.

• Bias error is the degree to which a model ts the training data. Variance error describes how much a
model’s results change in response to new data from valida on and test samples. Base error is due to
randomness in the data. Out-of-sample error equals bias error plus variance error plus base error.

• K-fold cross-valida on is a technique for mi ga ng the holdout sample problem (excessive reduc on
of the training set size). The data (excluding test sample and fresh data) are shu ed randomly and
then divided into k equal sub-samples, with k – 1 samples used as training samples and one sample,
the kth, used as a valida on sample.

• Regulariza on describes methods that reduce sta s cal variability in high-dimensional data
es ma on or predic on problems via reducing model complexity.

• LASSO (least absolute shrinkage and selec on operator) is a popular type of penalized regression
where the penalty term involves summing the absolute values of the regression coe cients. The
greater the number of included features, the larger the penalty. So, a feature must make a su cient
contribu on to model t to o set the penalty from including it.
fi
ti
ti
tt
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
tt
ff
fi
fi
ti
tti
ti
ti
fi
fi
ti
ti
ti
ti
ti
ti
ti
fl
ti
ffl
ti
fi
ffi
ti
ti
ti
ti
ti
ffi
fi
ti
ti
ti
 • Support vector machine (SVM) is a classi er that aims to seek the op mal hyperplane—the one that
separates the two sets of data points by the maxi- mum margin (and thus is typically used for
classi ca on).

• K-nearest neighbor (KNN) is a supervised learning technique most o en used for classi ca on. The
idea is to classify a new observa on by nding similari es (“nearness”) between it and its k-nearest
neighbors in the exist- ing dataset.

• Classi ca on and regression tree (CART) can be applied to predict either a categorical target variable,
producing a classi ca on tree, or a con nuous target variable, producing a regression tree.

• A binary CART is a combina on of an ini al root node, decision nodes, and terminal nodes. The root
node and each decision node represent a single feature (f) and a cuto value (c) for that feature. The
CART algorithm itera vely par ons the data into sub-groups un l terminal nodes are formed that
contain the predicted label.

• Ensemble learning is a technique of combining the predic ons from a collec on of models. It typically
produces more accurate and more stable predic ons than any single model.

• A random forest classi er is a collec on of many di erent decision trees generated by a bagging
method or by randomly reducing the number of features available during training.

• Principal components analysis (PCA) is an unsupervised ML algorithm that reduces highly correlated
features into fewer uncorrelated composite variables by transforming the feature covariance matrix.
PCA produces eigenvectors that de ne the principal components (i.e., the new uncorrelated
composite variables) and eigenvalues, which give the propor on of total variance in the ini al data
that is explained by each eigenvector and its associated principal component.
• K-means is an unsupervised ML algorithm that par ons observa ons into a xed number (k) of non-
overlapping clusters. Each cluster is characterized by its centroid, and each observa on belongs to the
cluster with the centroid to which that observa on is closest.

• Hierarchical clustering is an unsupervised itera ve algorithm that is used to build a hierarchy of

clusters. Two main strategies are used to de ne the intermediary clusters (i.e., those clusters
between the ini al dataset and the nal set of clustered data).

• Agglomera ve (bo om-up) hierarchical clustering begins with each observa on being its own cluster.
Then, the algorithm nds the two closest clusters, de ned by some measure of distance, and
combines them into a new, larger cluster. This process is repeated un l all observa ons are clumped
into a single cluster.

• Divisive (top-down) hierarchical clustering starts with all observa ons belonging to a single cluster.
The observa ons are then divided into two clusters based on some measure of distance. The
algorithm then progressively par ons the intermediate clusters into smaller clusters un l each
cluster contains only one observa on.

• Neural networks consist of nodes connected by links. They have three types of layers: an input layer,
hidden layers, and an output layer. Learning takes place in the hidden layer nodes, each of which
consists of a summa on operator and an ac va on func on. Neural networks have been successfully
applied to a variety of investment tasks characterized by non-lineari es and complex interac ons
among variables.
fi
fi
ti
ti
ti
ti
ti
fi
tt
ti
fi
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
fi
ti
fi
ti
fi
ti
fi
ti
fi
ti
ti
ti
ti
ti
ff
ti
fi
ti
ti
ti
ti
ti
ti
ti
ft
ti
ti
ti
ff
ti
ti
fi
ti
ti
fi
ti
ti
ti
ti
 • Neural networks with many hidden layers (at least 2 but o en more than 20) are known as deep
neural networks (DNNs) and are the backbone of the ar cial intelligence revolu on.

• Reinforcement learning (RL) involves an agent that should perform ac ons that will maximize its
rewards over me, taking into considera on the constraints of its environment.
ti
ti
ti
fi
ft
ti
ti
 Learning Module 7 Big Data Projects

In this reading, we have discussed the major steps in big data projects involving the development of
machine learning (ML) models—namely, those combining textual big data with structured inputs.

• Big data—de ned as data with volume, velocity, variety, and poten ally lower veracity—has
tremendous poten al for various ntech applica ons, including several related to investment
management.

• The main steps for tradi onal ML model building are conceptualiza on of the problem, data
collec on, data prepara on and wrangling, data explora on, and model training.

• For textual ML model building, the rst four steps di er somewhat from those used in the tradi onal
model: Text problem formula on, text cura on, text prepara on and wrangling, and text explora on
are typically necessary.

• For structured data, data prepara on and wrangling entail data cleansing and data preprocessing.
Data cleansing typically involves resolving incompleteness errors, invalidity errors, inaccuracy errors,
inconsistency errors, non-uniformity errors, and duplica on errors.

• Preprocessing for structured data typically involves performing the following transforma ons:
extrac on, aggrega on, ltra on, selec on, and conversion.

• Prepara on and wrangling text (unstructured) data involves a set of text-speci c cleansing and
preprocessing tasks. Text cleansing typically involves removing the following: html tags, punctua ons,
most numbers, and white spaces.

• Text preprocessing requires performing normaliza on that involves the following: lowercasing,
removing stop words, stemming, lemma za on, crea ng bag-of-words (BOW) and n-grams, and
organizing the BOW and n-grams into a document term matrix (DTM).

• Data explora on encompasses exploratory data analysis, feature selec on, and feature engineering.
Whereas histograms, box plots, and sca erplots are common techniques for exploring structured
data, word clouds are an e ec ve way to gain a high-level picture of the composi on of textual
content. These visualiza on tools help share knowledge among the team (business subject ma er
experts, quants, technologists, etc.) to help derive op mal solu ons.

• Feature selec on methods used for text data include term frequency, document frequency, chi-
square test, and a mutual informa on measure. Feature engineering for text data includes conver ng
numbers into tokens, crea ng n-grams, and using name en ty recogni on and parts of speech to
engineer new feature variables.

• The model training steps (method selec on, performance evalua on, and model tuning) o en do not
di er much for structured versus unstructured data projects.

• Model selec on is governed by the following factors: whether the data project involves labeled data
(supervised learning) or unlabeled data (unsupervised learning); the type of data (numerical,
con nuous, or categorical; text data; image data; speech data; etc.); and the size of the dataset.
ff
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
fi
ti
ff
ti
ti
ti
ti
fi
ti
fi
ti
tt
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ft
tt
ti
ti
ti
ti
 • Model performance evalua on involves error analysis using confusion matrixes, determining receiver
opera ng characteris cs, and calcula ng root mean square error.

• To carry out an error analysis for each model, a confusion matrix is created; true posi ves (TPs), true
nega ves (TNs), false posi ves (FPs), and false nega ves (FNs) are determined. Then, the following
performance metrics are calculated: accuracy, F1 score, precision, and recall. The higher the accuracy
and F1 score, the be er the model performance.

• To carry out receiver opera ng characteris c (ROC) analysis, ROC curves and area under the curve
(AUC) of various models are calculated and com- pared. The more convex the ROC curve and the
higher the AUC, the be er the model performance.

• Model tuning involves managing the trade-o between model bias error, associated with
under ng, and model variance error, associated with over ng. A ng curve of in-sample
(training sample) error and out-of-sample (cross-valida on sample) error on the y-axis versus model
complexity on the x-axis is useful for managing the bias vs. variance error trade-o .

• In a real-world big data project involving text data analysis for classifying and predic ng sen ment of
nancial text for par cular stocks, the text data are transformed into structured data for popula ng
the DTM, which is then used as the input for the ML algorithm.

• To derive term frequency (TF) at the sentence level and TF–IDF, both of which can be inputs to the
DTM, the following frequency measures should be used to create a term frequency measures table:
TotalWordsInSentence; TotalWordCount; TermFrequency (Collec on Level); WordCountInSentence;
SentenceCountWithWord; Document Frequency; and Inverse Document Frequency.
fi
ti
ti
fi
tti
ti
tt
ti
tt
ti
ti
ti
ti
ti
ff
ti
ti
fi
tti
ti
fi
tti
ff
ti
ti
ti
ti
 Economics
Learning Module 1 Currency Exchange Rates: Understanding Equilibrium Value

Exchange rates are among the most di cult nancial market prices to understand and therefore to
value. There is no simple, robust framework that investors can rely on in assessing the appropriate level
and likely movements of exchange rates.

Most economists believe that there is an equilibrium level or a path to that equilibrium value that a
currency will gravitate toward in the long run. Although short- and medium-term cyclical devia ons from
the long-run equilibrium path can be sizable and persistent, fundamental forces should eventually drive
the currency back toward its long-run equilibrium path. Evidence suggests that misalignments tend to
build up gradually over me. As these misalignments build, they are likely to generate serious economic
imbalances that will eventually lead to correc on of the underlying exchange rate misalignment.

We have described how changes in monetary policy, scal policy, current account trends, and capital
ows a ect exchange rate trends, as well as what role government interven on and capital controls can
play in counterac ng poten ally undesirable exchange rate movements. We have made the following
key points:

• Spot exchange rates apply to trades for the next se lement date (usually T + 2) for a given currency
pair. Forward exchange rates apply to trades to be se led at any longer maturity.

• Market makers quote bid and o er prices (in terms of the price currency) at which they will buy or
sell the base currency.
The o er price is always higher than the bid price.
The counterparty that asks for a two-sided price quote has the op on (but not the obliga on) to
deal at either the bid or o er price quoted.
The bid–o er spread depends on (1) the currency pair involved, (2) the me of day, (3) market
vola lity, (4) the transac on size, and (5) the rela onship between the dealer and the client.
Spreads are ghtest in highly liquid currency pairs, when the key market centers are open, and
when market vola lity is rela vely low.
• Absence of arbitrage requires the following:
The bid (o er) shown by a dealer in the interbank market cannot be higher (lower) than the
current interbank o er (bid) price.
The cross-rate bids (o ers) posted by a dealer must be lower (higher) than the implied cross-rate
o ers (bids) available in the interbank mar- ket. If they are not, then a triangular arbitrage
opportunity arises.

• Forward exchange rates are quoted in terms of points to be added to the spot exchange rate. If the
points are posi ve (nega ve), the base currency is trading at a forward premium (discount). The
points are propor onal to the interest rate di eren al and approximately propor onal to the me to
maturity.
fl
ff
ti
ff
ff
ff
ff
ti
ti
ti
ti
ti
ff
ff
ti
ti
ti
ff
ti
ti
ff
ffi
fi
ff
ti
ti
tt
ti
tt
fi
ti
ti
ti
ti
ti
ti
ti
 • Interna onal parity condi ons show us how expected in a on, interest rate di eren als, forward
exchange rates, and expected future spot exchange rates are linked. In an ideal world,
rela ve expected in a on rates should determine rela ve nominal interest rates,
rela ve interest rates should determine forward exchange rates, and
forward exchange rates should correctly an cipate the path of the future spot exchange rate.

• Interna onal parity condi ons tell us that countries with high (low) expected in a on rates should
see their currencies depreciate (appreciate) over me, that high-yield currencies should depreciate
rela ve to low-yield currencies over me, and that forward exchange rates should func on as
unbiased predictors of future spot exchange rates.

• With the excep on of covered interest rate parity, which is enforced by arbitrage, the key
interna onal parity condi ons rarely hold in either the short or medium term. However, the parity
condi ons tend to hold over rela vely long horizons.

• According to the theory of covered interest rate parity, a foreign-currency-denominated money

market investment that is completely hedged against exchange rate risk in the forward market should
yield exactly the same return as an otherwise iden cal domes c money market investment.

• According to the theory of uncovered interest rate parity, the expected change in a domes c
currency’s value should be fully re ected in domes c–foreign interest rate spreads. Hence, an
unhedged foreign-currency-denominated money market investment is expected to yield the same
return as an otherwise iden cal domes c money market investment.

• According to the ex ante purchasing power parity condi on, expected changes in exchange rates
should equal the di erence in expected na onal in a on rates.

• If both ex ante purchasing power parity and uncovered interest rate parity held, real interest rates
across all markets would be the same. This result is real interest rate parity.

• The interna onal Fisher e ect says that the nominal interest rate di eren al between two currencies
equals the di erence between the expected in a on rates. The interna onal Fisher e ect assumes
that risk premiums are the same throughout the world.

• If both covered and uncovered interest rate parity held, then forward rate parity would hold and the
market would set the forward exchange rate equal to the expected spot exchange rate: The forward
exchange rate would serve as an unbiased predictor of the future spot exchange rate.

• Most studies have found that high-yield currencies do not depreciate and low-yield currencies do not
appreciate as much as yield spreads would suggest over short to medium periods, thus viola ng the
theory of uncovered interest rate parity.

• Carry trades overweight high-yield currencies at the expense of low-yield currencies. Historically,
carry trades have generated a rac ve returns in benign market condi ons but tend to perform
poorly (i.e., are subject to crash risk) when market condi ons are highly vola le.

• According to a balance of payments approach, countries that run persistent current account de cits
will generally see their currencies weaken over me. Similarly, countries that run persistent current
account surpluses will tend to see their currencies appreciate over me.
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
ff
fl
ti
ti
ti
ti
ff
ti
tt
ti
fl
ti
ti
ti
ti
ti
fl
ti
ti
ti
ti
ti
fl
ti
ti
ti
fl
ti
ti
ti
ti
ff
ti
ti
ti
ti
ff
fl
ti
ti
ff
ti
ti
ti
fi
 • Large current account imbalances can persist for long periods of me before they trigger an
adjustment in exchange rates.

• Greater nancial integra on of the world’s capital markets and greater freedom of capital to ow
across na onal borders have increased the importance of global capital ows in determining
exchange rates.

• Countries that ins tute rela vely ght monetary policies, introduce structural economic reforms, and
lower budget de cits will o en see their currencies strengthen over me as capital ows respond
posi vely to rela vely high nominal interest rates, lower in a on expecta ons, a lower risk premium,
and an upward revision in the market’s assessment of what exchange rate level cons tutes long-run
fair value.

• Monetary policy a ects the exchange rate through a variety of channels. In the Mundell–Fleming
model, it does so primarily through the interest rate sensi vity of capital ows, strengthening the
currency when monetary pol- icy is ghtened and weakening it when monetary policy is eased. The
more sensi ve capital ows are to the change in interest rates, the greater the exchange rate’s
responsiveness to the change in monetary policy.

• In the monetary model of exchange rate determina on, monetary policy is deemed to have a direct
impact on the actual and expected path of in a on, which, via purchasing power parity, translates
into a corresponding impact on the exchange rate.

• Countries that pursue overly easy monetary policies will see their currencies depreciate over me.

• In the Mundell–Fleming model, an expansionary scal policy typically results in a rise in domes c
interest rates and an increase in economic ac vity. The rise in domes c interest rates should induce a
capital in ow, which is posi ve for the domes c currency, but the rise in economic ac vity should
contribute to a deteriora on of the trade balance, which is nega ve for the domes c currency. The
more mobile capital ows are, the greater the likelihood that the induced in ow of capital will
dominate the deteriora on in trade.

• Under condi ons of high capital mobility, countries that simultaneously pursue expansionary scal
policies and rela vely ght monetary policies should see their currencies strengthen over me.

• The por olio balance model of exchange rate determina on asserts that increases in government
debt resul ng from a rising budget de cit will be willingly held by investors only if they are
compensated in the form of a higher expected return. The higher expected return could come from
(1) higher interest rates and/or a higher risk premium, (2) deprecia on of the currency to a level
su cient to generate an cipa on of gains from subsequent currency apprecia on, or (3) some
combina on of the two.

• Surges in capital in ows can fuel boom-like condi ons, asset price bubbles, and currency
overvalua on.

• Many consider capital controls to be a legi mate part of a policymaker’s toolkit. The IMF believes that
capital controls may be needed to prevent exchange rates from overshoo ng, asset price bubbles
from forming, and future nancial condi ons from deteriora ng.
ffi
ti
tf
fi
ti
fl
ti
ti
ti
ti
ti
ti
fi
ti
ti
ff
fl
fl
ti
fl
ti
ti
ti
ti
fi
ft
ti
ti
ti
ti
ti
fi
ti
ti
fl
ti
ti
ti
fi
ti
ti
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
fl
fl
ti
ti
fl
ti
ti
fl
ti
ti
ti
fl
ti
fi
ti
 • The evidence indicates that government policies have had a signi cant impact on the course of
exchange rates. Rela ve to developed countries, emerging markets may have greater success in
managing their exchange rates because of their large foreign exchange reserve holdings, which
appear sizable rela ve to the limited turnover of FX transac ons in many emerging markets.

• Although each currency crisis is dis nct in some respects, the following factors were iden ed in one
or more studies:
1. Prior to a currency crisis, the capital markets have been liberalized to allow the free ow of capital.
2. There are large in ows of foreign capital (rela ve to GDP) in the period leading up to a crisis, with
short-term funding denominated in a foreign currency being par cularly problema c.
3. Currency crises are o en preceded by (and o en coincide with) banking crises.
4. Countries with xed or par ally xed exchange rates are more suscep ble to currency crises than
countries with oa ng exchange rates.
5. Foreign exchange reserves tend to decline precipitously as a crisis approaches.
6. In the period leading up to a crisis, the currency has risen substan ally rela ve to its historical
mean.
7. The terms of trade (exports rela ve to imports) o en deteriorate before a crisis.
8. Broad money growth and the ra o of M2 (a measure of money supply) to bank reserves tend to
rise prior to a crisis.
9. In a on tends to be signi cantly higher in pre-crisis periods compared with tranquil periods.
fl
ti
fl
fi
fl
ti
ti
ti
ft
fi
ti
ti
ti
fi
ti
ft
ti
ft
ti
ti
fi
ti
ti
ti
ti
fl
ti
fi
 Learning Module 2 Economic Growth

This reading focuses on the factors that determine the long-term growth trend in the economy. As part
of the development of global por olio equity and xed-income strategies, investors must be able to
determine both the near-term and the sustainable rates of growth within a country. Doing so requires
iden fying and forecas ng the factors that determine the level of GDP and that determine long-term
sustainable trends in economic growth.

• The sustainable rate of economic growth is measured by the rate of increase in the economy’s
produc ve capacity or poten al GDP.

• Growth in real GDP measures how rapidly the total economy is expanding. Per capita GDP, de ned as
real GDP divided by popula on, measures the standard of living in each country.

• The growth rate of real GDP and the level of per capita real GDP vary widely among countries. As a
result, investment opportuni es di er by country.

• Equity markets respond to an cipated growth in earnings. Higher sustain- able economic growth
should lead to higher earnings growth and equity market valua on ra os, all other things being
equal.

• In the long run, the growth rate of earnings cannot exceed the growth in poten al GDP. Labor
produc vity is cri cal because it a ects the level of the upper limit. A permanent increase in
produc vity growth will raise the upper limit on earnings growth and should translate into faster
long-run earnings growth and a corresponding increase in stock price apprecia on.

• For global xed-income investors, a cri cal macroeconomic variable is the rate of in a on. One of the
best indicators of short- to intermediate-term in a on trends is the di erence between the growth
rate of actual and poten al GDP.
• Capital deepening, an increase in the capital-to-labor ra o, occurs when the growth rate of capital
(net investment) exceeds the growth rate of labor. In a graph of output per capita versus the capital-
to-labor ra o, it is re ected by a move along the curve (i.e., the produc on func on).

• An increase in total factor produc vity causes a propor onal upward shi in the en re produc on
func on.

• One method of measuring sustainable growth uses the produc on func on and the growth
accoun ng framework developed by Solow. It arrives at the growth rate of poten al GDP by
es ma ng the growth rates of the economy’s capital and labor inputs plus an es mate of total factor
produc vity.

• An alterna ve method measures poten al growth as the long-term growth rate of the labor force
plus the long-term growth rate of labor produc vity.

• The forces driving economic growth include the quan ty and quality of labor and the supply of non-
ICT and ICT capital, public capital, raw materials, and technological knowledge.

• The labor supply is determined by popula on growth, the labor force par cipa on rate, and net
immigra on. The physical capital stock in a country increases with net investment. The correla on
between long-run economic growth and the rate of investment is high.
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
fl
ti
ti
ti
ti
ti
ti
tf
ti
ff
ff
ti
ti
ti
ti
fl
fi
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ft
ti
ti
ti
ti
ti
ti
ti
ti
fl
ti
fi
ti
ti
 • Technological advances are discoveries that make it possible to produce more or higher-quality goods
and services with the same resources or inputs. Technology is a major factor determining TFP. TFP is
the main factor a ec ng long-term, sustainable economic growth rates in developed countries and
also includes the cumula ve e ects of scien c advances, applied research and development,
improvements in management methods, and ways of organizing produc on that raise the produc ve
capacity of factories and o ces.

• Total factor produc vity, es mated using a growth accoun ng equa on, is the residual component of
growth a er accoun ng for the weighted contribu ons of all explicit factors (e.g., labor and capital).

• Labor produc vity is de ned as output per worker or per hour worked. Growth in labor produc vity
depends on capital deepening and technological progress.

• The academic growth literature is divided into three theories —the classical view, the neoclassical
model, and the new endogenous growth view.

• In the classical model, growth in per capita income is only temporary because an exploding
popula on with limited resources brings per capita income growth to an end.

• In the neoclassical model, a sustained increase in investment increases the economy’s growth rate
only in the short run. Capital is subject to diminishing marginal returns, so long-run growth depends
solely on popula on growth, progress in TFP, and labor’s share of income.

• The neoclassical model assumes that the produc on func on exhibits diminishing marginal
produc vity with respect to any individual input.

• The point at which capital per worker and output per worker are growing at equal, sustainable rates is
called the steady state or balanced growth path for the economy. In the steady state, total output
grows at the rate of labor force growth plus the rate of growth of TFP divided by the elas city of
output with respect to labor input.

• The following parameters a ect the steady-state values for the capital-to-labor ra o and output per
worker: saving rate, labor force growth, growth in TFP, deprecia on rate, and elas city of output with
respect to capital.

• The main cri cism of the neoclassical model is that it provides no quan able predic on of the rate
or form of TFP change. TFP progress is regarded as exogenous to the model.

• Endogenous growth theory explains technological progress within the model rather than trea ng it as
exogenous. As a result, self-sustaining growth emerges as a natural consequence of the model and
the economy does not converge to a steady-state rate of growth that is independent of saving/
investment decisions.

• Unlike the neoclassical model, where increasing capital will result in diminishing marginal returns, the
endogenous growth model allows for the possibility of constant or even increasing returns to capital
in the aggregate economy.
ti
ti
ft
ti
ti
ti
ff
ti
ti
ti
fi
ti
ffi
ti
ff
ff
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
 • In the endogenous growth model, expenditures made on R&D and for human capital may have large
posi ve externali es or spillover e ects. Private spending by companies on knowledge capital
generates bene ts to the economy as a whole that exceed the private bene t to the company.

• The convergence hypothesis predicts that the rates of growth of produc vity and GDP should be
higher in the developing countries. Those higher growth rates imply that the per capita GDP gap
between developing and developed economies should narrow over me. The evidence on
convergence is mixed.

• Countries fail to converge because of low rates of investment and savings, lack of property rights,
poli cal instability, poor educa on and health, restric ons on trade, and tax and regulatory policies
that discourage work and inves ng.

• Opening an economy to nancial and trade ows has a major impact on economic growth. The
evidence suggests that more open and trade-oriented economies will grow at a faster rate.
ti
ti
fi
ti
fi
ti
ti
ff
fl
ti
ti
ti
fi
 Learning Module 3 Economics of Regula on

Knowledge of regula on is important because regula on has poten ally far-reaching and signi cant
e ects. These e ects can range from macro-level e ects on the economy to micro-level e ects on
individual en es and securi es.

Regula on originates from a variety of sources and in a variety of areas. A frame- work that includes
types of regulators and regula on as well as areas of regula on that may a ect the en ty of interest
(including the economy as an en ty) is useful. The framework will help in assessing possible e ects of
new regula on. It can also help in assessing the e ects of regula on on various en es.

More than one regulator may develop regula ons in response to a par cular issue. Each of the relevant
regulators may have di erent objec ves and choose to address the issue using di erent regulatory tools.
In developing regula ons, the regulator should consider costs and bene ts. In the analysis, the net
regulatory burden (private costs less private bene ts of regula on) may also be relevant. Poten al costs
and bene ts, regardless of the perspec ve, may be di cult to assess. A cri cal aspect of regulatory
analysis, however, is assessing the costs and bene ts of regula on.

The following are some key points of the reading.

• The existence of informa onal fric ons and externali es creates a need for regula on. Regula on is
expected to have societal bene ts and should be assessed using cost–bene t analysis.

• The regula on of securi es markets and nancial ins tu ons is extensive and complex because of the
consequences of failures in the nancial system. These consequences include nancial losses, loss of
con dence, and disrup on of commerce.

• The focus of regulators in nancial markets includes pruden al supervision, nancial stability, market
integrity, and economic growth.

• Regulatory compe on is compe on among di erent regulatory bodies to use regula on in order to
a ract certain en es.

• The breadth of regula on of commerce necessitates the use of a framework that iden es poten al
areas of regula on. This framework can be referenced to iden fy speci c areas of regula on, both
exis ng and an cipated, that may a ect the en ty of interest.

• Legisla ve bodies, regulatory bodies, and courts typically enact regula on.

• Regulatory bodies include government agencies and independent regulators granted authority by a
government or governmental agency. Some independent regulators are self-regula ng organiza ons.

• Typically, legisla ve bodies enact broad laws or statutes. Regulatory bodies issue administra ve
regula ons, o en implemen ng statutes. Courts interpret statutes and administra ve regula ons;
these interpreta ons may result in judicial law.

• Interdependence in the ac ons and poten ally con ic ng objec ves of regulators is an important
considera on for regulators, regulated en es, and those assessing the e ects of regula on.

• Regula on that arises to enhance the interests of regulated en es re ects regulatory capture.
tt
ff
fi
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ft
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
fi
ti
ti
ti
ti
fi
fi
ti
ti
ti
ti
ti
ff
ti
ti
fi
ti
ti
ti
ti
ti
ff
ff
fi
fi
ff
fl
ti
ti
ti
ffi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fl
ti
fi
fi
ff
ti
ff
fi
fi
fi
ff
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ff
ti
ti
ff
ti
fi
ti
ti
ti
 • Regulators have responsibility for both substan ve and procedural laws. The former focuses on rights
and responsibili es of en es and rela onships among en es. The la er focuses on the protec on
and enforcement of the former.

• Regulatory arbitrage is the use of regula on by an en ty to exploit di erences in economic substance

and regulatory interpreta on or in regulatory regimes to the en ty’s bene t.

• There are many regulatory tools available to regulators, including regulatory mandates and
restric ons on behaviors, provision of public goods, and pub- lic nancing of private projects.

• The choice of regulatory tool should be consistent with maintaining a stable regulatory environment.
“Stable” does not mean unchanging but, rather, refers to desirable a ributes of regula on, including
predictability, e ec ve- ness in achieving objec ves, me consistency, and enforceability.
• In assessing regula on and regulatory outcomes, regulators should conduct ongoing cost–bene t
analyses, develop techniques to enhance the measurement of these outcomes, and use economic
principles to guide them.

• Net regulatory burden to the en ty of interest is an important considera on for analysts.

ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
tt
ff
tt
ti
fi
ti
fi
ti
 Financial Statement Analysis
Learning Module 1 Intercorporate Investments

Intercompany investments play a signi cant role in business ac vi es and create signi cant challenges
for the analyst in assessing company performance. Investments in other companies can take ve basic
forms: investments in nancial assets, investments in associates, joint ventures, business combina ons,
and investments in special purpose and variable interest en es. Key concepts are as follows:

• Investments in nancial assets are those in which the investor has no signi cant in uence. They can
be measured and reported as
Fair value through pro t or loss.
Fair value through other comprehensive income.
Amor zed cost.

• IFRS and US GAAP treat investments in nancial assets in a similar manner.

• Investments in associates and joint ventures are those in which the investor has signi cant in uence,
but not control, over the investee’s business ac vi es. Because the investor can exert signi cant
in uence over nancial and opera ng policy decisions, IFRS and US GAAP require the equity method
of accoun ng because it provides a more objec ve basis for repor ng investment income.
The equity method requires the investor to recognize income as earned rather than when
dividends are received.
The equity investment is carried at cost, plus its share of post-acquisi on income (a er
adjustments) less dividends received.
The equity investment is reported as a single line item on the balance sheet and on the income
statement.

• IFRS and US GAAP accoun ng standards require the use of the acquisi on method to account for
business combina ons. Fair value of the considera on given is the appropriate measurement for
iden able assets and liabili es acquired in the business combina on.

• Goodwill is the di erence between the acquisi on value and the fair value of the target’s iden able
net tangible and intangible assets. Because it is considered to have an inde nite life, it is not
amor zed. Instead, it is evaluated at least annually for impairment. Impairment losses are reported
on the income statement. IFRS use a one-step approach to determine and measure the impairment
loss, whereas US GAAP uses a two-step approach.

• If the acquiring company acquires less than 100%, non-controlling (minority) shareholders’ interests
are reported on the consolidated nancial statements. IFRS allows the non-controlling interest to be
measured at either its fair value (full goodwill) or at the non-controlling interest’s propor onate share
of the acquiree’s iden able net assets (par al goodwill). US GAAP requires the non-controlling
interest to be measured at fair value (full goodwill).
fl
ti
ti
ti
fi
ti
fi
fi
ff
ti
ti
fi
fi
fi
ti
ti
fi
ti
fi
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
fi
fl
ft
fi
fi
ti
fi
fi
fl
ti
fi
ti
 • Consolidated nancial statements are prepared in each repor ng period.

• Special purpose (SPEs) and variable interest en es (VIEs) are required to be consolidated by the
en ty which is expected to absorb the majority of the expected losses or receive the majority of
expected residual bene ts.
ti
fi
fi
ti
ti
ti
 Learning Module 2 Employee Compensa on: Post-Employment and Share-Based

This reading discussed two di erent forms of employee compensa on: post-employment bene ts and
share-based compensa on. Although di erent, the two are similar in that they are forms of
compensa on outside of the standard salary arrangements. They also involve complex valua on,
accoun ng, and repor ng issues. Although IFRS and US GAAP are converging on accoun ng and
repor ng, it is important to note that di erences in a country’s social system, laws, and regula ons can
result in di erences in a company’s pension and share-based compensa on plans that may be re ected
in the company’s earnings and nancial reports.

Key points include the following:

• De ned contribu on pension plans specify (de ne) only the amount of contribu on to the plan; the
eventual amount of the pension bene t to the employee will depend on the value of an employee’s
plan assets at the me of re rement.

• Balance sheet repor ng is less analy cally relevant for de ned contribu on plans because companies
make contribu ons to de ned contribu on plans as the expense arises and thus no liabili es accrue
for that type of plan.

• De ned bene t pension plans specify (de ne) the amount of the pension bene t, o en determined
by a plan formula, under which the eventual amount of the bene t to the employee is a func on of
length of service and nal salary.

• De ned bene t pension plan obliga ons are funded by the sponsoring company contribu ng assets
to a pension trust, a separate legal en ty. Di erences exist in countries’ regulatory requirements for
companies to fund de ned bene t pension plan obliga ons.

• Both IFRS and US GAAP require companies to report on their balance sheet a pension liability or asset
equal to the projected bene t obliga on minus the fair value of plan assets. The amount of a pension
asset that can be reported is subject to a ceiling.

• Under IFRS, the components of periodic pension cost are recognised as follows: Service cost is
recognised in P&L, net interest income/expense is recognised in P&L, and remeasurements are
recognised in OCI and are not amor sed to future P&L.

• Under US GAAP, the components of periodic pension cost recognised in P&L include current service
costs, interest expense on the pension obliga on, and expected returns on plan assets (which reduces
the cost). Other components of periodic pension cost—including past service costs, actuarial gains
and losses, and di erences between expected and actual returns on plan assets—are recognised in
OCI and amor sed to future P&L.

• Es mates of the future obliga on under de ned bene t pension plans and other post-employment
bene ts are sensi ve to numerous assump ons, including discount rates, assumed annual
compensa on increases, expected return on plan assets, and assumed health care cost in a on.

• Employee compensa on packages are structured to ful ll varied objec ves, including sa sfying
employees’ needs for liquidity, retaining employees, and providing incen ves to employees.
ti
fi
fi
fi
fi
ti
ti
ti
ti
ff
fi
fi
ti
ti
ti
ti
ff
ti
ti
ti
fi
fi
ti
ti
fi
ti
fi
ff
ti
fi
fi
ti
ti
ti
ti
ti
fi
ti
ff
ff
ti
fi
ti
fi
ff
ti
fi
fi
ti
fi
fi
fi
ti
ti
ti
ti
ti
fi
ti
ft
ti
ti
ti
ti
fl
ti
ti
ti
ti
fi
fl
 • Common components of employee compensa on packages are salary, bonuses, and share-based
compensa on.

• Share-based compensa on serves to align employees’ interests with those of the shareholders. It
includes stocks and stock op ons.

• Share-based compensa on has the advantage of requiring no current-period cash outlays.

• Share-based compensa on expense is reported at fair value under IFRS and US GAAP.

• The valua on technique, or op on pricing model, that a company uses is an important choice in
determining fair value and is disclosed.

• Key assump ons and input into op on pricing models include such items as exercise price, stock price
vola lity, es mated life of each award, es mated number of op ons that will be forfeited, dividend
yield, and the risk-free rate of interest. Certain assump ons are highly subjec ve, such as stock price
vola lity or the expected life of stock op ons, and can greatly change the es mated fair value and
thus compensa on expense.
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
 Learning Module 3 Mul na onal Opera ons

The transla on of foreign currency amounts is an important accoun ng issue for companies with
mul na onal opera ons. Foreign exchange rate uctua ons cause the func onal currency values of
foreign currency assets and liabili es resul ng from foreign currency transac ons as well as from foreign
subsidiaries to change over me. These changes in value give rise to foreign exchange di erences that
companies’ nancial statements must re ect. Determining how to measure these foreign exchange
di erences and whether to include them in the calcula on of net income are the major issues in
accoun ng for mul na onal opera ons.

• The local currency is the na onal currency of the country where an en ty is located. The func onal
currency is the currency of the primary economic environment in which an en ty operates. Normally,
the local currency is an en ty’s func onal currency. For accoun ng purposes, any currency other than
an en ty’s func onal currency is a foreign currency for that en ty. The currency in which nancial
statement amounts are presented is known as the presenta on currency. In most cases, the
presenta on currency will be the same as the local currency.

• When an export sale (import purchase) on an account is denominated in a foreign currency, the sales
revenue (inventory) and foreign currency account receivable (account payable) are translated into the
seller’s (buyer’s) func onal currency using the exchange rate on the transac on date. Any change in
the func onal currency value of the foreign currency account receivable (account payable) that
occurs between the transac on date and the se lement date is recognized as a foreign currency
transac on gain or loss in net income.

• If a balance sheet date falls between the transac on date and the se lement date, the foreign
currency account receivable (account payable) is translated at the exchange rate at the balance sheet
date. The change in the func onal currency value of the foreign currency account receivable (account
pay- able) is recognized as a foreign currency transac on gain or loss in income. Analysts should
understand that these gains and losses are unrealized at the me they are recognized and might or
might not be realized when the transac ons are se led.

• A foreign currency transac on gain arises when an en ty has a foreign currency receivable and the
foreign currency strengthens or it has a foreign currency payable and the foreign currency weakens. A
foreign currency transac on loss arises when an en ty has a foreign currency receivable and the
foreign currency weakens or it has a foreign currency payable and the foreign currency strengthens.

• Companies must disclose the net foreign currency gain or loss included in income. They may choose
to report foreign currency transac on gains and losses as a component of opera ng income or as a
component of non-opera ng income. If two companies choose to report foreign currency transac on
gains and losses di erently, opera ng pro t and opera ng pro t margin might not be directly
comparable between the two companies.

• To prepare consolidated nancial statements, foreign currency nancial statements of foreign

opera ons must be translated into the parent company’s presenta on currency. The major
conceptual issues related to this transla on process are, What is the appropriate exchange rate for
transla ng each nancial statement item, and how should the resul ng transla on adjustment be
re ected in the consolidated nancial statements? Two di er- ent transla on methods are used
worldwide.
ff
fl
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
fi
ff
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
fl
ti
fi
ti
tt
ti
fl
tt
ti
ti
ti
ti
ti
ti
ff
ti
ti
fi
ti
ti
fi
ti
ti
ti
tt
ti
ti
ti
ti
ti
ti
ti
ti
ff
fi
ti
ti
 • Under the current rate method, assets and liabili es are translated at the current exchange rate,
equity items are translated at historical exchange rates, and revenues and expenses are translated at
the exchange rate that existed when the underlying transac on occurred. For prac cal reasons, an
average exchange rate is o en used to translate income items.

• Under the temporal method, monetary assets (and non-monetary assets measured at current value)
and monetary liabili es (and non-monetary liabili es measured at current value) are translated at the
current exchange rate. Non-monetary assets and liabili es not measured at current value and equity
items are translated at historical exchange rates. Revenues and expenses, other than those expenses
related to non-monetary assets, are translated at the exchange rate that existed when the underlying
transac on occurred. Expenses related to non-monetary assets are translated at the exchange rates
used for the related assets.

• Under both IFRS and US GAAP, the func onal currency of a foreign opera on determines the method
to be used in transla ng its foreign currency nancial statements into the parent’s presenta on
currency and whether the resul ng transla on adjustment is recognized in income or as a separate
component of equity.

• The foreign currency nancial statements of a foreign opera on that has a foreign currency as its
func onal currency are translated using the current rate method, and the transla on adjustment is
accumulated as a separate component of equity. The cumula ve transla on adjustment related to a
speci c foreign en ty is transferred to net income when that en ty is sold or otherwise disposed of.
The balance sheet risk exposure associated with the current rate method is equal to the foreign
subsidiary’s net asset posi on.

• The foreign currency nancial statements of a foreign opera on that has the parent’s presenta on
currency as its func onal currency are translated using the temporal method, and the transla on
adjustment is included as a gain or loss in income. US GAAP refer to this process as remeasurement.
The balance sheet exposure associated with the temporal method is equal to the foreign subsidiary’s
net monetary asset/liability posi on (adjusted for non-monetary items measured at current value).

• IFRS and US GAAP di er with respect to the transla on of foreign currency nancial statements of
foreign opera ons located in a highly in a onary country. Under IFRS, the foreign currency
statements are rst restated for local in a on and then translated using the current exchange rate.
Under US GAAP, the foreign currency nancial statements are translated using the temporal method,
with no restatement for in a on.

• Applying di erent transla on methods for a given foreign opera on can result in very di erent
amounts reported in the parent’s consolidated nancial statements.

• Companies must disclose the total amount of transla on gain or loss reported in income and the
amount of transla on adjustment included in a separate component of stockholders’ equity.
Companies are not required to separately disclose the component of transla on gain or loss arising
from foreign currency transac ons and the component arising from applica on of the temporal
method.

• Disclosures related to transla on adjustments reported in equity can be used to include these as
gains and losses in determining an adjusted amount of income following a clean-surplus approach to
income measurement.
ti
fi
ti
ff
ti
fi
ti
ti
ti
ti
ti
ff
fi
fi
ti
ti
fl
ft
ti
ti
ti
ti
ti
fi
fl
ti
fl
ti
ti
ti
fi
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ff
ti
ti
ti
 • Foreign currency transla on rules are well established in both IFRS and US GAAP. Fortunately, except
for the treatment of foreign opera ons located in highly in a onary countries, the two sets of
standards have no major di erences in this area. The ability to understand the impact of foreign
currency transla on on the nancial results of a company using IFRS should apply equally well in the
analysis of nancial statements prepared in accordance with US GAAP.

• An analyst can obtain informa on about the tax impact of mul na onal opera ons from companies’
disclosure on e ec ve tax rates.

• For a mul na onal company, sales growth is driven not only by changes in volume and price but also
by changes in the exchange rates between the repor ng currency and the currency in which sales are
made. Arguably, growth in sales that comes from changes in volume or price is more sustain- able
than growth in sales that comes from changes in exchange rates.
ti
fi
ti
ff
ti
ti
ti
ff
fi
ti
ti
ti
fl
ti
ti
ti
ti
 Learning Module 4 Analysis of Financial Ins tu ons

• Financial ins tu ons’ systemic importance results in heavy regula on of their ac vi es.

• Systemic risk refers to the risk of impairment in some part of the nancial system that then has the
poten al to spread throughout other parts of the nancial system and thereby to nega vely a ect
the en re economy.

• The Basel Commi ee, a standing commi ee of the Bank for Interna onal Se lements, includes
representa ves from central banks and bank supervisors from around the world.

• The Basel Commi ee’s interna onal regulatory framework for banks includes minimum capital
requirements, minimum liquidity requirements, and stable funding requirements.

• Among the interna onal organiza ons that focus on nancial stability are the Financial Stability
Board, the Interna onal Associa on of Insurance Supervisors, the Interna onal Associa on of
Deposit Insurers, and the Interna onal Organiza on of Securi es Commissions.

• Another dis nc ve feature of nancial ins tu ons (compared to manufacturing or merchandising

companies) is that their produc ve assets are predominantly nancial assets, such as loans and
securi es, crea ng greater direct exposures to a variety of risks, such as credit risk, liquidity risk,
market risk, and interest rate risk. In general, the values of their assets are rela vely close to fair
market values.

• A widely used approach to analyzing a bank, CAMELS, considers a bank’s Capital adequacy, Asset
quality, Management capabili es, Earnings su ciency, Liquidity posi on, and Sensi vity to market
risk.

• “Capital adequacy,” described in terms of the propor on of the bank’s assets that is funded with
capital, indicates that a bank has enough capital to absorb poten al losses without severely damaging
its nancial posi on.

• “Asset quality” includes the concept of quality of the bank’s assets— credit quality and diversi ca on
—and the concept of overall sound risk management.

• “Management capabili es” refers to the bank management’s ability to iden fy and exploit
appropriate business opportuni es and to simultaneously manage associated risks.

• “Earnings” refers to the bank’s return on capital rela ve to cost of capital and also includes the
concept of earnings quality.

• “Liquidity” refers to the amount of liquid assets held by the bank rela ve to its near-term expected
cash ows. Under Basel III, liquidity also refers to the stability of the bank’s funding sources.

• “Sensi vity to market risk” pertains to how adverse changes in markets (including interest rate,
exchange rate, equity, and commodity markets) could a ect the bank’s earnings and capital posi on.

• In addi on to the CAMELS components, important a ributes deserving analysts’ a en on include

government support, the banking en ty’s mis- sion, corporate culture and compe ve environment,
o -balance-sheet items, segment informa on, currency exposure, and risk disclosures.
ff
fi
fl
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
tt
tt
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
tt
ti
ti
ti
ti
ffi
ti
ti
fi
ti
tt
ti
fi
ff
ti
fi
ti
fi
ti
ti
ti
ti
ti
ti
tt
ti
ti
ti
tt
ti
ti
ti
ti
ti
ti
fi
ff
ti
ti
 • Insurance companies are typically categorized as property and casualty (P&C) or life and health (L&H).

• Insurance companies earn revenues from premiums (amounts paid by the purchaser of insurance
products) and from investment income earned on the oat (amounts collected as premiums and not
yet paid out as bene ts).

• P&C insurers’ policies are usually short term, and the nal cost will usually be known within a year of
a covered event, whereas L&H insurers’ policies are usually longer term. P&C insurers’ claims are
more variable, whereas L&H insurers’ claims are more predictable.

• For both types of insurance companies, important areas for analysis include business pro le, earnings
characteris cs, investment returns, liquidity, and capitaliza on. In addi on, analysis of P&C
companies’ pro tability includes analysis of loss reserves and the combined ra o.
ti
fi
fi
fi
fl
ti
ti
ti
fi
 Learning Module 5 Evalua ng Quality of Financial Reports

Assessing the quality of nancial reports—both repor ng quality and results quality—is an important
analy cal skill.

• The quality of nancial repor ng can be thought of as spanning a con nuum from the highest quality
to the lowest.

• Poten al problems that a ect the quality of nancial repor ng broadly include revenue and expense
recogni on on the income statement; classi ca on on the statement of cash ows; and the
recogni on, classi ca on, and measurement of assets and liabili es on the balance sheet.

• Typical steps involved in evalua ng nancial repor ng quality include an understanding of the
company’s business and industry in which the company is opera ng; comparison of the nancial
statements in the current period and the previous period to iden fy any signi cant di erences in line
items; an evalua on of the company’s accoun ng policies, especially any unusual revenue and
expense recogni on compared with those of other companies in the same industry; nancial ra o
analysis; examina on of the statement of cash ows with par cular focus on di erences between net
income and opera ng cash ows; perusal of risk disclosures; and review of management
compensa on and insider transac ons.

• High-quality earnings increase the value of the company more than low-quality earnings, and the
term “high-quality earnings” assumes that repor ng quality is high.

• Low-quality earnings are insu cient to cover the company’s cost of capital and/or are derived from
non-recurring, one-o ac vi es. In addi on, the term “low-quality earnings” can be used when the
reported informa on does not provide a useful indica on of the company’s performance.

• Various alterna ves have been used as indicators of earnings quality: recur- ring earnings, earnings
persistence and related measures of accruals, bea ng benchmarks, and a er-the-fact con rma ons
of poor-quality earnings, such as enforcement ac ons and restatements.

• Earnings that have a signi cant accrual component are less persistent and thus may revert to the
mean more quickly.

• A company that consistently reports earnings that exactly meet or only narrowly beat benchmarks
can raise ques ons about its earnings quality.

• Cases of accoun ng malfeasance have commonly involved issues with revenue recogni on, such as
premature recogni on of revenues or the recogni on of fraudulent revenues.

• Cases of accoun ng malfeasance have involved misrepresenta on of expenditures as assets rather

than as expenses or misrepresenta on of the ming or amount of expenses.

• Bankruptcy predic on models, used in assessing nancial results quality, quan fy the likelihood that a
company will default on its debt and/or declare bankruptcy.
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ff
ti
fi
ti
fi
ff
ti
fl
ti
ti
ffi
ti
ti
ti
fi
ti
fi
fi
ti
ti
fl
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ft
fi
fl
ti
ff
fi
ff
ti
fi
fi
ti
ti
 • Similar to the term “earnings quality,” when reported cash ows are described as being high quality, it
means that the company’s underlying economic performance was sa sfactory in terms of increasing
the value of the rm, and it also implies that the company had high repor ng quality (i.e., that the
informa on calculated and disclosed by the company was a good re ec on of economic reality). Cash
ow can be described as “low quality” either because the reported informa on properly represents
genuinely bad economic performance or because the reported informa on misrepresents economic
reality.

• For the balance sheet, high nancial repor ng quality is indicated by completeness, unbiased
measurement, and clear presenta on.

• A balance sheet with signi cant amounts of o -balance-sheet debt would lack the completeness
aspect of nancial repor ng quality.

• Unbiased measurement is a par cularly important aspect of nancial repor ng quality for assets and
liabili es for which valua on is subjec ve.

• A company’s nancial statements can provide useful indicators of nancial or opera ng risk.

• The management commentary (also referred to as the management discussion and analysis, or
MD&A) can give users of the nancial statements informa on that is helpful in assessing the
company’s risk exposures and approaches to managing risk.

• Required disclosures regarding, for example, changes in senior management or inability to make a
mely ling of required nancial reports can be a warning sign of problems with nancial repor ng
quality.

• The nancial press can be a useful source of informa on about risk when, for example, a nancial
reporter uncovers nancial repor ng issues that had not previously been recognized. An analyst
should undertake addi onal inves ga on of any issue iden ed.
fl
ti
fi
ti
fi
ti
fi
fi
fi
fi
ti
fi
ti
ti
fi
fi
fi
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
fl
fi
fi
fi
fl
ti
ti
ti
ti
ti
ti
fi
ti
fi
ti
 Learning Module 6 Integra on of Financial Statement Analysis Techniques

The case study demonstrates the use of a nancial analysis framework in investment decision making.
Although each analysis undertaken may have a di erent focus, purpose, and context that result in the
applica on of di erent techniques and tools, the case demonstrates the use of a common nancial
statement analysis framework. The analyst starts with a global, summarized view of a company and its
a ributes and digs below the surface of the nancial statements to nd economic truths that are not
apparent from a super cial review. In the case of Nestlé, the analyst applied disaggrega on techniques
to review the company’s performance in terms of ROE and then successively examined the drivers of
ROE in increasing detail to evaluate management’s skills in capital alloca on.

An economic decision is reached, which is consistent with the primary reason for performing nancial
analysis: to facilitate an economic decision.
tt
ti
ff
fi
ti
fi
fi
ff
fi
ti
ti
fi
fi
 Corporate Issuers
Learning Module 1 Financial Statement Modeling

Industry and company analysis are essen al tools of fundamental analysis. The key points made include
the following:

• Analysts can use a top-down, bo om-up, or hybrid approach to forecas ng income and expenses.
Top-down approaches usually begin at the level of the overall economy. Bo om-up approaches begin
at the level of the individual company or unit within the company (e.g., business segment). Time-
series approaches are considered bo om-up, although me-series analysis can be a tool used in top-
down approaches. Hybrid approaches include elements of top-down and bo om-up approaches.

• In a “growth rela ve to GDP growth” approach to forecas ng revenue, the analyst forecasts the
growth rate of nominal GDP and industry and company growth rela ve to GDP growth.

• In a “market growth and market share” approach to forecas ng revenue, the analyst combines
forecasts of growth in par cular markets with forecasts of a company’s market share in the individual
markets.

• Opera ng margins that are posi vely correlated with sales provide evidence of economies of scale in
an industry.

• Some balance sheet line items, such as retained earnings, ow directly from the income statement,
whereas accounts receivable, accounts payable, and inventory are very closely linked to income
statement projec ons.

• A common way to model working capital accounts is to use e ciency ra os.

• ROIC, de ned as net opera ng pro t less adjusted taxes divided by the di erence between opera ng
assets and opera ng liabili es, is an a er-tax measure of pro tability. High and persistent levels of
ROIC are o en associated with having a compe ve advantage.

• Compe ve factors a ect a company’s ability to nego ate lower input prices with suppliers and to
raise prices for products and services. Porter’s ve forces framework can be used as a basis for
iden fying such factors.

• In a on (de a on) a ects pricing strategy depending on industry structure, compe ve forces, and
the nature of consumer demand.

• When a technological development results in a new product that threatens to cannibalize demand for
an exis ng product, a unit forecast for the new product combined with an expected cannibaliza on
factor can be used to es mate the impact on future demand for the exis ng product.
fl
ti
ti
ti
ti
ti
ti
fi
ft
fl
ti
ti
ti
ti
ff
ff
ti
ti
ti
ti
ti
tt
fi
tt
ft
ti
fi
ti
ti
ti
ti
ti
fl
ti
fi
ffi
ti
ti
ti
ti
ff
tt
tt
ti
ti
ti
ti
 • Factors in uencing the choice of the explicit forecast horizon include the projected holding period, an
investor’s average por olio turnover, the cyclicality of an industry, company-speci c factors, and
employer preferences.

• Key behavioral biases that in uence analyst forecasts are overcon dence, illusion of control
conserva sm, representa veness, and con rma on bias.
ti
fl
tf
ti
fl
fi
ti
fi
fi
 Learning Module 2 Analysis of Dividends and Share Repurchases

A company’s cash dividend payment and share repurchase policies cons tute its payout policy. Both
entail the distribu on of the company’s cash to its shareholders a ect the form in which shareholders
receive the return on their investment. Among the points this reading has made are the following:

• Dividends can take the form of regular or irregular cash payments, stock dividends, or stock splits.
Only cash dividends are payments to shareholders. Stock dividends and splits merely carve equity
into smaller pieces and do not create wealth for shareholders. Reverse stock splits usually occur a er
a stock has dropped to a very low price and do not a ect shareholder wealth.

• Regular cash dividends—unlike irregular cash dividends, stock splits, and stock dividends—represent
a commitment to pay cash to stockholders on a quarterly, semiannual, or annual basis.

• There are three general theories on investor preference for dividends. The rst, MM, argues that
given perfect markets dividend policy is irrelevant. The second, “bird in hand” theory, contends that
investors value a dollar of dividends today more than uncertain capital gains in the future. The third
theory argues that in countries in which dividends are taxed at higher rates than capital gains, taxable
investors prefer that companies reinvest earnings in pro table growth opportuni es or repurchase
shares so they receive more of the return in the form of capital gains.

• An argument for dividend irrelevance given perfect markets is that corporate dividend policy is
irrelevant because shareholders can create their preferred cash ow stream by selling the company’s
shares (“homemade dividends”).

• Dividend declara ons may provide informa on to current and prospec ve shareholders regarding
management’s con dence in the prospects of the company. Ini a ng a dividend or increasing a
dividend sends a posi ve signal, whereas cu ng a dividend or omi ng a dividend typically sends a
nega ve signal. In addi on, some ins tu onal and individual shareholders see regular cash dividend
payments as a measure of investment quality.

• Payment of dividends can help reduce the agency con icts between man- agers and shareholders,
but it also can worsen con icts of interest between shareholders and debtholders.

• Empirically, several factors appear to in uence dividend policy, including investment opportuni es for
the company, the vola lity expected in its future earnings, nancial exibility, tax considera ons,
ota on costs, and contractual and legal restric ons.

• Under double taxa on systems, dividends are taxed at both the corporate and shareholder level.
Under tax imputa on systems, a shareholder receives a tax credit on dividends for the tax paid on
corporate pro ts. Under split-rate taxa on systems, corporate pro ts are taxed at di erent rates
depending on whether the pro ts are retained or paid out in dividends.

• Companies with outstanding debt o en are restricted in the amount of dividends they can pay
because of debt covenants and legal restric ons. Some ins tu ons require that a company pay a
dividend to be on their “approved” investment list. If a company funds capital expenditures by
borrowing while paying earnings out in dividends, it will incur ota on costs on new debt issues.
fl
ti
ti
fi
ti
ti
ti
ti
fi
ti
ti
ti
fl
fi
ft
ti
ti
fl
ti
ti
ti
tti
ti
ff
fl
fi
ti
fi
ti
fl
ti
fl
ti
ff
fi
ti
tti
fl
ti
ti
fi
ti
ff
ti
ti
ft
 • Using a stable dividend policy, a company tries to align its dividend growth rate to the company’s
long-term earnings growth rate. Dividends may increase even in years when earnings decline, and
dividends will increase at a lower rate than earnings in boom years.

• A stable dividend policy can be represented by a gradual adjustment process in which the expected
dividend is equal to last year’s dividend per share plus [(Expected earnings × Target payout ra o −
Previous dividend) × Adjustment factor].

• Using a constant dividend payout ra o policy, a company applies a target dividend payout ra o to
current earnings; therefore, dividends are more vola le than with a stable dividend policy.

• Share repurchases, or buybacks, most o en occur in the open market. Alterna vely, tender o ers
occur at a xed price or at a price range through a Dutch auc on. Shareholders who do not tender
increase their rela ve posi on in the company. Direct nego a ons with major shareholders to get
them to sell their posi ons are less common because they could destroy value for remaining
stockholders.

• Share repurchases made with excess cash have the poten al to increase earnings per share, whereas
share repurchases made with borrowed funds can increase, decrease, or not a ect earnings per share
depending on the company’s a er-tax borrowing rate and earnings yield.

• A share repurchase is equivalent to the payment of a cash dividend of equal amount in its e ect on
total shareholders’ wealth, all other things being equal.

• If the buyback market price per share is greater (less) than the book value per share, then the book
value per share will decrease (increase).

• Companies can repurchase shares in lieu of increasing cash dividends. Share repurchases usually o er
company management more exibility than cash dividends by not establishing the expecta on that a
par cular level of cash distribu on will be maintained.

• Companies can pay regular cash dividends supplemented by share repurchases. In years of
extraordinary increases in earnings, share repurchases can subs tute for special cash dividends.

• On the one hand, share repurchases can signal that company o cials think their shares are
undervalued. On the other hand, share repurchases could send a nega ve signal that the company
has few posi ve NPV opportuni es.

• Analysts are interested in how safe a company’s dividend is, speci cally whether the company’s
earnings and, more importantly, its cash ow are su cient to sustain the payment of the dividend.

• Early warning signs of whether a company can sustain its dividend include the dividend coverage
ra o, the level of dividend yield, whether the company borrows to pay the dividend, and the
company’s past dividend record.
ti
ti
fi
ti
ti
ti
ti
fl
ft
ti
ti
ti
ft
fl
ffi
ti
ti
ti
ti
ti
ffi
ti
fi
ti
ti
ff
ti
ff
ti
ti
ff
ff
 Learning Module 3 ESG Considera ons in Investment Analysis

• Shareholder ownership structures are commonly classi ed as dispersed, concentrated, or a hybrid of

the two.

• Dispersed ownership re ects the existence of many shareholders, none of which, either individually
or collec vely, has the ability to exercise control over the corpora on. Concentrated corporate
ownership re ects an individual shareholder or a group (controlling shareholders) with the ability to
exercise control over the corpora on.

• Controlling shareholders may be either majority shareholders or minority shareholders.

• Horizontal ownership involves companies with mutual business interests that have cross-holding
share arrangements with each other. Ver cal (or pyramid) ownership involves a company or group
that has a controlling interest in two or more holding companies, which in turn have controlling
interests in various opera ng companies.

• Dual-class (or mul ple-class) shares grant one or more share classes superior or even sole vo ng
rights while other share classes have inferior or no vo ng rights.

• Types of in uen al owners include banks, families, sovereign governments, ins tu onal investors,
group companies, private equity rms, foreign investors, managers, and board directors.

• A corpora on’s board of directors is typically structured as either one er or two er. A one- er board
consists of a single board of directors, composed of execu ve (internal) and non-execu ve (external)
directors. A two- er board consists of a supervisory board that oversees a management board. CEO
duality exists when the chief execu ve o cer also serves as chairperson of the board.

• A primary challenge of integra ng ESG factors into investment analysis is iden fying and obtaining
informa on that is relevant, comparable, and decision-useful.

• ESG informa on and metrics are inconsistently reported by companies, and such disclosure is
voluntary, which provides addi onal challenges for analysts. In an ESG context, materiality typically
refers to ESG-related issues that are expected to a ect a company’s opera ons or nancial
performance and the valua on of its securi es.

• Corporate governance considera ons, such as the structure of the board of directors, tend to be
reasonably consistent across most companies. In contrast, environmental and social considera ons
o en di er greatly.

• Analysts typically use three main sources of informa on to iden fy a company’s (or industry’s) ESG
factors: (1) proprietary research, (2) ra ngs and analysis from ESG data providers, or (3) research from
not-for-pro t industry organiza ons and ini a ves.

• In equity analysis, ESG integra on is used to both iden fy poten al opportuni es and mi gate
downside risk, whereas in xed-income analysis, ESG integra on is generally focused on mi ga ng
downside risk.

• A typical star ng point for ESG integra on is the iden ca on of material qualita ve and quan ta ve
ESG factors that pertain to a company or its industry.
ft
ff
ti
ti
ti
fl
fi
ti
ti
fl
ti
ti
ti
fl
ti
fi
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ffi
ti
ti
ti
ti
ff
ti
ti
ti
ti
fi
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
 Learning Module 5 Corporate Restructuring

• Corporate issuers seek to alter their des ny, as described by the corporate life cycle, by taking ac ons
known as restructurings.

• Restructurings include investment ac ons that increase the size and scope of an issuer’s business,
divestment ac ons that decrease size or scope, and restructuring ac ons that do not a ect scope but
improve performance.

• Investment ac ons include equity investments, joint ventures, and acquisi ons. Investment ac ons
are o en made by issuers seeking growth, synergies, or undervalued targets.

• Divestment ac ons include sales and spin o s and are made by issuers seeking to increase growth or
pro tability or reduce risk by shedding certain divisions and assets.

• Restructuring ac ons, including cost cu ng, balance sheet restructurings, and reorganiza ons, do
not change the size or scope of issuers but are aimed at improving returns on capital to historical or
peer levels.

• The evalua on of a corporate restructuring is composed of four phases: ini al evalua on, preliminary
evalua on, modeling, and upda ng the investment thesis. The en re evalua on is generally done
only for material restructurings.

• The ini al evalua on of a corporate restructuring answers the following ques ons: What is
happening? When is it happening? Is it material? And why is it happening?

• Materiality is de ned by both size and t. One rule of thumb for size is that large ac ons are those
that are greater than 10% of an issuer’s enterprise value (e.g., for an acquisi on, considera on in
excess of 10% of the acquirer’s pre-announcement enterprise value). Fit refers to the alignment
between the ac on and an analyst’s expecta ons for the issuer.

• Three common valua on methods for companies involved in corporate restructurings, during the
preliminary valua on phase of the evalua on, are comparable company, comparable transac on, and
premium paid analysis.

• Corporate restructurings must be modeled on the nancial statements based on the situa onal
speci cs. Es mated nancial statements that include the e ect of a restructuring are known as pro
forma nancial statements.

• The weighted average cost of capital for an issuer is determined by the weights of di erent capital
types and the cons tuent costs of capital. The costs of capital are in uenced by both bo om-up and
top-down drivers. Bo om-up drivers include stability, pro tability, leverage, and asset speci city.
Corporate restructurings a ect the cost of capital by a ec ng these drivers.
fi
ft
fi
ti
fi
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
fi
tt
ti
ff
ti
ti
fi
tti
ti
ti
ff
ti
fi
ff
fi
ti
ff
ti
fl
ti
ti
ti
ti
ti
ti
ti
ff
ti
ff
tt
ti
ti
ti
fi
ti
ti
ti
 Equity Valuation
Learning Module 1 Equity Valua on: Applica ons and Processes

In this reading, we have discussed the scope of equity valua on, outlined the valua on process,
introduced valua on concepts and models, discussed the analyst’s role and responsibili es in conduc ng
valua on, and described the elements of an e ec ve research report in which analysts communicate
their valua on analysis.

• Valua on is the es ma on of an asset’s value based on either variables perceived to be related to

future investment returns or comparisons with closely similar assets.

• The intrinsic value of an asset is its value given a hypothe cally complete understanding of the asset’s
investment characteris cs.

• The assump on that the market price of a security can diverge from its intrinsic value—as suggested
by the ra onal e cient markets formula on of e cient market theory—underpins ac ve inves ng.

• Intrinsic value incorporates the going-concern assump on, that is, the assump on that a company
will con nue opera ng for the foreseeable future. In contrast, liquida on value is the company’s
value if it were dis- solved and its assets sold individually.

• Fair value is the price at which an asset (or liability) would change hands if neither buyer nor seller
were under compulsion to buy/sell and both were informed about material underlying facts.

• In addi on to stock selec on by ac ve traders, valua on is also used for:

inferring (extrac ng) market expecta ons;
evalua ng corporate events;
issuing fairness opinions;
evalua ng business strategies and models; and
appraising private businesses.

• The valua on process has ve steps:

1. Understanding the business.
2. Forecas ng company performance.
3. Selec ng the appropriate valua on model.
4. Conver ng forecasts to a valua on.
5. Applying the analy cal results in the form of recommenda ons and conclusions.

• Understanding the business includes evalua ng industry prospects, compe ve posi on, and
corporate strategies—all of which contribute to making more accurate forecasts. Understanding the
business also involves analysis of nancial reports, including evalua ng the quality of a company’s
earnings.
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ffi
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
fi
ti
ti
ti
ti
ti
ff
ffi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
 • In forecas ng company performance, a top-down forecas ng approach moves from macroeconomic
forecasts to industry forecasts and then to individual company and asset forecasts. A bo om-up
forecas ng approach aggregates individual company forecasts to industry forecasts, which in turn
may be aggregated to macroeconomic forecasts.

• Selec ng the appropriate valua on approach means choosing an approach that is:
consistent with the characteris cs of the company being valued;
appropriate given the availability and quality of the data; and
consistent with the analyst’s valua on purpose and perspec ve.

• Two broad categories of valua on models are absolute valua on models and rela ve valua on
models.
Absolute valua on models specify an asset’s intrinsic value, supplying a point es mate of value
that can be compared with market price. Present value models of common stock (also called
discounted cash ow models) are the most important type of absolute valua on model.
Rela ve valua on models specify an asset’s value rela ve to the value of another asset. As applied
to equity valua on, rela ve valua on is also known as the method of comparables, which involves
comparison of a stock’s price mul ple to a benchmark price mul ple. The benchmark price
mul ple can be based on a similar stock or on the average price mul ple of some group of stocks.

• Two important aspects of conver ng forecasts to valua on are sensi vity analysis and situa onal
adjustments.
Sensi vity analysis is an analysis to determine how changes in an assumed input would a ect the
outcome of an analysis.
Situa onal adjustments include control premiums (premiums for a controlling interest in the
company), discounts for lack of marketability (discounts re ec ng the lack of a public market for
the company’s shares), and illiquidity discounts (discounts re ec ng the lack of a liquid market for
the company’s shares).

• Applying valua on conclusions depends on the purpose of the valua on.

• In performing valua ons, analysts must hold themselves accountable to both standards of
competence and standards of conduct.

• An e ec ve research report:
contains mely informa on;
is wri en in clear, incisive language;
is objec ve and well researched, with key assump ons clearly iden ed;
dis nguishes clearly between facts and opinions;
contains analysis, forecasts, valua on, and a recommenda on that are internally consistent;
presents su cient informa on that the reader can cri que the valua on;
states the risk factors for an investment in the company; and
discloses any poten al con icts of interest faced by the analyst.

• Analysts have an obliga on to provide substan ve and meaningful con- tent. CFA Ins tute members
have an addi onal overriding responsibility to adhere to the CFA Ins tute Code of Ethics and relevant
speci c Standards of Professional Conduct.
ti
ti
ti
ff
ti
fi
ti
ti
tt
ti
ti
ti
ti
ti
ffi
ti
ti
ti
ti
ti
fl
ti
ti
ti
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fl
ti
fl
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
tt
ff
ti
ti
 Learning Module 2 Discounted Dividend Valua on

We have provided an overview of DCF models of valua on, discussed the es ma on of a stock’s required
rate of return, and presented in detail the dividend discount model.

• In DCF models, the value of any asset is the present value of its (expected) future cash ows where

V0 is the value of the asset as of t = 0 (today), CFt is the (expected)

cash ow at me t, and r is the discount rate or required rate of
return. For in nitely lived assets such as common stocks, n runs to
in nity.

• Several alterna ve streams of expected cash ows can be used to value equi es, including dividends,
free cash ow, and residual income. A discounted dividend approach is most suitable for dividend-
paying stocks in which the company has a discernible dividend policy that has an understandable
rela onship to the company’s pro tability and the investor has a non-control (minority ownership)
perspec ve.

• The free cash ow approach (FCFF or FCFE) might be appropriate when the company does not pay
dividends, dividends di er substan ally from FCFE, free cash ows align with pro tability, or the
investor takes a control (majority ownership) perspec ve.

• The residual income approach can be useful when the company does not pay dividends (as an
alterna ve to a FCF approach) or free cash ow is nega ve.

• The DDM with a single holding period gives stock value as

where D1 is the expected dividend at Time 1

and V0 is the stock’s (expected) value at Time
0. Assuming that V0 is equal to today’s
market price, P0, the
expected holding-period return is

• The expression for the DDM for any given nite holding period n and the general expression for the
DDM are, respec vely,
fi
ti
fl
ti
ti
fl
ti
fi
fl
ti
ti
ff
fi
ti
fi
fl
fl
ti
ti
ti
ti
fl
ti
ti
ti
fi
fl
 • There are two main approaches to the problem of forecas ng dividends. First, an analyst can assign
the en re stream of expected future dividends to one of several stylized growth pa erns. Second, an
analyst can forecast a nite number of dividends individually up to a terminal point and value the
remaining dividends either by assigning them to a stylized growth pa ern or by forecas ng share
price as of the terminal point of the dividend forecasts.

• The Gordon growth model assumes that dividends grow at a constant rate g forever, so that
Dt = Dt–1(1 + g). The dividend stream in the Gordon growth model has a value of

• The value of non-callable xed-rate perpetual preferred stock is V0 = D/r, where D is the stock’s
(constant) annual dividend.

• Assuming that price equals value, the Gordon growth model es mate of a stock’s expected rate of
return is

• Given an es mate of the next-period dividend and the stock’s required rate of return, the Gordon
growth model can be used to es mate the dividend growth rate implied by the current market price
(making a constant growth rate assump on).

• The present value of growth opportuni es is the part of a stock’s total value, V0, that comes from
pro table future growth opportuni es in contrast to the value associated with assets already in place.
The rela onship is V0 = E1/r + PVGO, where E1/r is de ned as the no-growth value per share.

• The leading price-to-earnings ra o (P0/E1) and the trailing price-to-earnings ra o (P0/E0) can be
expressed in terms of the Gordon growth model as, respec vely,

The foregoing expressions give a stock’s jus ed price-to-earnings ra o based on forecasts of

fundamentals (given that the Gordon growth model is appropriate).
fi
ti
ti
ti
fi
fi
ti
ti
ti
ti
ti
ti
fi
fi
ti
ti
ti
ti
tt
ti
tt
ti
 • The Gordon growth model may be useful for valuing broad-based equity indexes and the stock of
businesses with earnings that are expected to grow at a stable rate comparable to or lower than the
economy’s nominal growth rate.

• Gordon growth model values are very sensi ve to the assumed growth rate and required rate of
return.

• For many companies, growth falls into phases. In the growth phase, a company enjoys an abnormally
high growth rate in earnings per share, called supernormal growth. In the transi on phase, earnings
growth slows. In the mature phase, the company reaches an equilibrium in which such factors as
earnings growth and the return on equity stabilize at levels that can be sustained long term. Analysts
o en apply mul stage DCF models to value the stock of a company with mul stage growth prospects.

• The two-stage dividend discount model assumes di erent growth rates in Stage 1 and Stage 2:

where gS is the expected dividend growth rate in the rst period and gL is the expected growth rate in
the second period.

• The terminal stock value, Vn, is some mes found with the Gordon growth model or with some other
method, such as applying a P/E mul plier to fore- casted EPS as of the terminal date.

• The H-model assumes that the dividend growth rate declines linearly from a high supernormal rate to
the normal growth rate during Stage 1 and then grows at a constant normal growth rate therea er:

• There are two basic three-stage models. In one version, the growth rate in the middle stage is
constant. In the second version, the growth rate declines linearly in Stage 2 and becomes constant
and normal in Stage 3.

• In addi on to valuing equi es, the IRR of a DDM, assuming assets are correctly priced in the
marketplace, has been used to es mate required returns. For simpler models (such as the one-period
model, the Gordon growth model, and the H-model), well-known formulas may be used to calculate
these rates of return. For many dividend streams, however, the rate of return must be found by trial
and error, producing a discount rate that equates the present value of the forecasted dividend stream
to the current market price.
ft
ti
ti
ti
ti
ti
ti
ti
ff
fi
ti
ti
ft
 • Mul stage DDM models can accommodate a wide variety of pa erns of expected dividends. Even
though such models may use stylized assump ons about growth, they can provide useful
approxima ons.

• Dividend growth rates can be obtained from analyst forecasts, sta s cal forecas ng models, or
company fundamentals. The sustainable growth rate depends on the ROE and the earnings reten on
rate, b: g = b × ROE. This
ti
ti
ti
tt
ti
ti
ti
ti
 Learning Module 3 Free Cash Flow Valua on

Discounted cash ow models are widely used by analysts to value companies.

• Free cash ow to the rm (FCFF) and free cash ow to equity (FCFE) are the cash ows available to,
respec vely, all of the investors in the company and to common stockholders.

• Analysts like to use free cash ow (either FCFF or FCFE) as the return
if the company is not paying dividends;
if the company pays dividends but the dividends paid di er signi cantly from the company’s
capacity to pay dividends;
if free cash ows align with pro tability within a reasonable forecast period with which the analyst
is comfortable; or
if the investor takes a control perspec ve.

• The FCFF valua on approach es mates the value of the rm as the present value of future FCFF
discounted at the weighted average cost of capital:

The value of equity is the value of the rm minus the value of the rm’s debt:
Equity value = Firm value – Market value of debt.

Dividing the total value of equity by the number of outstanding shares gives the value per share.
The WACC formula is

• The value of the rm if FCFF is growing at a constant rate is

• With the FCFE valua on approach, the value of equity can be found by dis- coun ng FCFE at the
required rate of return on equity, r:
Dividing the total value of equity by the number of outstanding
shares gives the value per share.
ti
fl
fl
ti
fl
fi
ti
fi
fl
fi
ti
ti
fi
ti
fl
ff
fi
fi
fi
ti
fl
 • The value of equity if FCFE is growing at a constant rate is

• FCFF and FCFE are frequently calculated by star ng with net income:
FCFF = NI + NCC + Int(1 – Tax rate) – FCInv – WCInv.
FCFE = NI + NCC – FCInv – WCInv + Net borrowing.

• FCFF and FCFE are related to each other as follows:

FCFE = FCFF – Int(1 – Tax rate) + Net borrowing.

• FCFF and FCFE can be calculated by star ng from cash ow from opera ons:
FCFF = CFO + Int(1 – Tax rate) – FCInv.
FCFE = CFO – FCInv + Net borrowing.

• FCFF can also be calculated from EBIT or EBITDA:

FCFF = EBIT(1 – Tax rate) + Dep – FCInv – WCInv.
FCFF = EBITDA(1 – Tax rate) + Dep(Tax rate) – FCInv – WCInv.
FCFE can then be found by using FCFE = FCFF – Int(1 – Tax rate) + Net borrowing.

• Finding CFO, FCFF, and FCFE may require careful interpreta on of corporate nancial statements. In
some cases, the necessary informa on may not be transparent.

• Earnings components such as net income, EBIT, EBITDA, and CFO should not be used as cash ow
measures to value a rm. These earnings components either double-count or ignore parts of the cash
ow stream.

• FCFF or FCFE valua on expressions can be easily adapted to accommodate complicated capital
structures, such as those that include preferred stock.

• A general expression for the two-stage FCFF valua on model is

• A general expression for the two-stage FCFE valua on model is

fl
ti
fi
ti
ti
ti
ti
ti
fl
ti
ti
fi
fl
 • One common two-stage model assumes a constant growth rate in each stage, and a second common
model assumes declining growth in Stage 1 followed by a long-run sustainable growth rate in Stage 2.

• To forecast FCFF and FCFE, analysts build a variety of models of varying complexity. A common
approach is to forecast sales, with pro tability, investments, and nancing derived from changes in
sales.

• Three-stage models are o en considered to be good approxima ons for cash ow streams that, in
reality, uctuate from year to year.

• Non-opera ng assets, such as excess cash and marketable securi es, noncurrent investment
securi es, and nonperforming assets, are usually segregated from the company’s opera ng assets.
They are valued separately and then added to the value of the company’s opera ng assets to nd
total rm value.
fi
ti
fl
ti
ft
fi
ti
ti
fi
fl
ti
ti
fi
 Learning Module 4 Market-Based Valua on: Price and Enterprise Value Mul ples

We have de ned and explained the most important valua on indicators in professional use and
illustrated their applica on to a variety of valua on problems.

• Price mul ples are ra os of a stock’s price to some measure of value per share.

• Price mul ples are most frequently applied to valua on in the method of comparables. This method
involves using a price mul ple to evaluate whether an asset is rela vely undervalued, fairly valued, or
overvalued in rela on to a benchmark value of the mul ple.

• The benchmark value of the mul ple may be the mul ple of a similar company or the median or
average value of the mul ple for a peer group of companies, an industry, an economic sector, an
equity index, or the company’s own median or average past values of the mul ple.

• The economic ra onale for the method of comparables is the law of one price.

• Price mul ples may also be applied to valua on in the method based on forecasted fundamentals.
Discounted cash ow (DCF) models provide the basis and ra onale for this method. Fundamentals
also interest analysts who use the method of comparables because di erences between a price
mul ple and its benchmark value may be explained by di erences in fundamentals.

• The key idea behind the use of price-to-earnings ra os (P/Es) is that earning power is a chief driver of
investment value and earnings per share (EPS) is probably the primary focus of security analysts’
a en on. The EPS gure, however, is frequently subject to distor on, o en vola le, and some mes
nega ve.

• The two alterna ve de ni ons of P/E are trailing P/E, based on the most recent four quarters of EPS,
and forward P/E, based on next year’s expected earnings.

• Analysts address the problem of cyclicality by normalizing EPS—that is, calcula ng the level of EPS
that the business could achieve currently under mid-cyclical condi ons (normalized EPS).

• Two methods to normalize EPS are the method of historical average EPS (calculated over the most
recent full cycle) and the method of average return on equity (EPS = average ROE mul plied by
current book value per share).

• Earnings yield (E/P) is the reciprocal of the P/E. When stocks have zero or nega ve EPS, a ranking by
earnings yield is meaningful whereas a ranking by P/E is not.

• Historical trailing P/Es should be calculated with EPS lagged a su cient amount of me to avoid look-
ahead bias. The same principle applies to other mul ples calculated on a trailing basis.

• The fundamental drivers of P/E are the expected earnings growth rate and the required rate of
return. The jus ed P/E based on fundamentals bears a posi ve rela onship to the rst factor and an
inverse rela onship to the second factor.

• The PEG (P/E-to-growth) ra o is a tool to incorporate the impact of earnings growth on P/E. The PEG
ra o is calculated as the ra o of the P/E to the consensus growth forecast. Stocks with low PEG ra os
are, all else equal, more a rac ve than stocks with high PEG ra os.
tt
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
fi
ti
fl
ti
fi
ti
fi
ti
ti
ti
tt
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ffi
ti
ti
ti
ti
ff
ft
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
 • We can es mate terminal value in mul stage DCF models by using price mul ples based on
comparables. The expression for terminal value, Vn, is (using P/E as the example)
Vn = Benchmark value of trailing P/E × En or
Vn = Benchmark value of forward P/E × En+1.

• Book value per share is intended to represent, on a per-share basis, the investment that common
shareholders have in the company. In a on, technological change, and accoun ng distor ons,
however, may impair the use of book value for this purpose.

• Book value is calculated as common shareholders’ equity divided by the number of shares
outstanding. Analysts adjust book value to accurately re ect the value of the shareholders’
investment and to make P/B (the price-to-book ra o) more useful for comparing di erent stocks.

• The fundamental drivers of P/B are ROE and the required rate of return. The jus ed P/B based on
fundamentals bears a posi ve rela onship to the rst factor and an inverse rela onship to the second
factor.

• An important ra onale for using the price-to-sales ra o (P/S) is that sales, as the top line in an income
statement, are generally less subject to distor on or manipula on than other fundamentals, such as
EPS or book value. Sales are also more stable than earnings and are never nega ve.

• P/S fails to take into account di erences in cost structure between businesses, may not properly
re ect the situa on of companies losing money, and may be subject to manipula on through
revenue recogni on prac ces.

• The fundamental drivers of P/S are pro t margin, growth rate, and the required rate of return. The
jus ed P/S based on fundamentals bears a posi ve rela onship to the rst two factors and an
inverse rela onship to the third factor.

• Enterprise value (EV) is total company value (the market value of debt, com- mon equity, and
preferred equity) minus the value of cash and investments.

• The ra o of EV to total sales is conceptually preferable to P/S because EV/S facilitates comparisons
among companies with varying capital structures.

• A key idea behind the use of price to cash ow is that cash ow is less subject to manipula on than
are earnings. Price-to-cash- ow mul ples are o en more stable than P/Es. Some common
approxima ons to cash ow from opera ons have limita ons, however, because they ignore items
that may be subject to manipula on.

• The major cash ow (and related) concepts used in mul ples are earnings plus non-cash charges (CF),
cash ow from opera ons (CFO), free cash ow to equity (FCFE), and earnings before interest, taxes,
deprecia on, and amor za on (EBITDA).

• In calcula ng price to cash ow, the earnings-plus-non cash-charges concept is tradi onally used,
although FCFE has the strongest link to nancial theory.

• CF and EBITDA are not strictly cash ow numbers because they do not account for non-cash revenue
and net changes in working capital.
fl
ti
fi
fl
ti
ti
ti
ti
ti
ti
ti
fl
ti
ti
ti
ti
fl
ti
ti
ti
fl
fl
ff
ti
ti
fl
ti
fl
ti
fi
fi
ti
ti
fl
fl
ti
ft
ti
fi
ti
ti
ti
fl
ti
ti
fl
ti
fi
ti
ti
ti
ti
ti
ti
fi
ff
ti
ti
ti
 • The fundamental drivers of price to cash ow, however de ned, are the expected growth rate of
future cash ow and the required rate of return. The jus ed price to cash ow based on
fundamentals bears a posi ve rela on- ship to the rst factor and an inverse rela onship to the
second.

• EV/EBITDA is preferred to P/EBITDA because EBITDA, as a pre-interest number, is a ow to all

providers of capital.

• EV/EBITDA may be more appropriate than P/E for comparing companies with di erent amounts of
nancial leverage (debt).

• EV/EBITDA is frequently used in the valua on of capital-intensive businesses.

• The fundamental drivers of EV/EBITDA are the expected growth rate in free cash ow to the rm,
pro tability, and the weighted average cost of capital. The jus ed EV/EBITDA based on fundamentals
bears a posi ve rela on- ship to the rst two factors and an inverse rela onship to the third.

• Dividend yield has been used as a valua on indicator because it is a component of total return and is
less risky than capital apprecia on.

• Trailing dividend yield is calculated as four mes the most recent quarterly per-share dividend divided
by the current market price.

• The fundamental drivers of dividend yield are the expected growth rate in dividends and the required
rate of return.

• Comparing companies across borders frequently involves dealing with di erences in accoun ng
standards, cultural di erences, economic di erences, and resul ng di erences in risk and growth
opportuni es.

• Momentum indicators relate either price or a fundamental to the me series of the price’s or
fundamental’s own past values (in some cases, to their expected values).

• Momentum valua on indicators include earnings surprise, standardized unexpected earnings (SUE),
and rela ve strength.

• Unexpected earnings (or earnings surprise) equals the di erence between reported earnings and
expected earnings.

• SUE is unexpected earnings divided by the standard devia on in past unexpected earnings.

• Rela ve-strength indicators allow comparison of a stock’s performance during a period either with its
own past performance ( rst type) or with the performance of some group of stocks (second type).
The ra onale for using rela ve strength is the thesis that pa erns of persistence or reversal in returns
exist.

• Screening is the applica on of a set of criteria to reduce an investment universe to a smaller set of
investments and is a part of many stock selec on disciplines. In general, limita ons of such screens
include the lack of control in vendor-provided data of the calcula on of important inputs and the
absence of qualita ve factors.
fi
fi
ti
ti
ti
ti
fl
ti
ti
ti
ti
ff
ti
fi
ti
ti
ti
ti
fi
ti
fl
ti
ti
ff
ti
fi
ti
ff
fi
ti
fi
tt
ti
fi
ti
ti
ti
ff
ti
ff
fl
ti
ff
ti
fl
fl
ti
fi
 Learning Module 5 Residual Income Valua on

We have discussed the use of residual income models in valua on. Residual income is an appealing
economic concept because it a empts to measure economic pro t, which are pro ts a er accoun ng
for all opportunity costs of capital.

• Residual income is calculated as net income minus a deduc on for the cost of equity capital. The
deduc on, called the equity charge, is equal to equity capital mul plied by the required rate of return
on equity (the cost of equity capital in percent).

• Economic value added (EVA) is a commercial implementa on of the residual income concept. EVA =
NOPAT − (C% × TC), where NOPAT is net opera ng pro t a er taxes, C% is the percent cost of capital,
and TC is total capital.

• Residual income models (including commercial implementa ons) are used not only for equity
valua on but also to measure internal corporate performance and for determining execu ve
compensa on.

• We can forecast per-share residual income as forecasted earnings per share minus the required rate
of return on equity mul plied by beginning book value per share. Alterna vely, per-share residual
income can be forecasted as beginning book value per share mul plied by the di erence between
forecasted ROE and the required rate of return on equity.

• In the residual income model, the intrinsic value of a share of common stock is the sum of book value
per share and the present value of expected future per-share residual income. In the residual income
model, the equivalent mathema cal expressions for intrinsic value of a common stock are
ti
ti
ti
ti
tt
ti
ti
ti
fi
ti
ft
ti
ti
ti
ti
fi
ti
ti
ff
fi
ft
ti
ti
 • In the two-stage model with con nuing residual income in stage two, the intrinsic value of a s_hare of
stock is

• In most cases, value is recognized earlier in the residual income model com- pared with other present
value models of stock value, such as the dividend discount model.

• Strengths of the residual income model include the following:

Terminal values do not make up a large por on of the value rela ve to other models.
The models use readily available accoun ng data.
The models can be used in the absence of dividends and near-term posi ve free cash ows.
The models can be used when cash ows are unpredictable.

• Weaknesses of the residual income model include the following:

The models are based on accoun ng data that can be subject to manipula on by management.
Accoun ng data used as inputs may require signi cant adjustments.
The models require that the clean surplus rela on holds, or that the analyst makes appropriate
adjustments when the clean surplus rela on does not hold.

• The residual income model is most appropriate in the following cases:

A company is not paying dividends or if it exhibits an unpredictable dividend pa ern.
A company has nega ve free cash ow many years out but is expected to generate posi ve cash
ow at some point in the future.
A great deal of uncertainty exists in forecas ng terminal values.

• The fundamental determinants or drivers of residual income are book value of equity and return on
equity.

• Residual income valua on is most closely related to P/B. When the present value of expected future
residual income is posi ve (nega ve), the jus ed P/B based on fundamentals is greater than (less
than) one.
fl
ti
ti
ti
ti
ti
ti
ti
fl
fl
ti
ti
ti
ti
ti
ti
fi
fi
ti
ti
ti
tt
fl
ti
 • When fully consistent assump ons are used to forecast earnings, cash ow, dividends, book value,
and residual income through a full set of pro forma (projected) nancial statements, and the same
required rate of return on equity is used as the discount rate, the same es mate of value should
result from a residual income, dividend discount, or free cash ow valua on. In prac ce, however,
analysts may nd one model easier to apply and possibly arrive at di erent valua ons using the
di erent models.

• Con nuing residual income is residual income a er the forecast horizon. Frequently, one of the
following assump ons concerning con nuing residual income is made:
Residual income con nues inde nitely at a posi ve level. (One varia on of this assump on is that
residual income con nues inde nitely at the rate of in a on, meaning it is constant in real terms.)
Residual income is zero from the terminal year forward.
Residual income declines to zero as ROE reverts to the cost of equity over me.
Residual income declines to some mean level.

• The residual income model assumes the clean surplus rela on of Bt = Bt–1 + Et − Dt. In other terms,
the ending book value of equity equals the beginning book value plus earnings minus dividends,
apart from ownership transac ons.

• In prac ce, to apply the residual income model most accurately, the analyst may need to do the
following:
adjust book value of common equity for:
• o -balance-sheet items;
• discrepancies from fair value; or
• the amor za on of certain intangible assets.
adjust reported net income to re ect clean surplus accoun ng.
adjust reported net income for non-recurring items misclassi ed as recurring items.
ff
ff
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
fi
fi
fl
ti
ti
ft
fl
ti
ti
ti
fi
fl
fi
ti
ff
fl
ti
ti
ti
ti
ti
ti
 Learning Module 6 Private Company Valua on

We have provided an overview of key elements of private company valua on and contrasted public and
private company valua ons.

• Company- and stock-speci c factors may in uence the selec on of appropriate valua on methods
and assump ons for private company valua ons. Stock-speci c factors may result in a lower value for
an equity interest in a private company rela ve to a public company.

• Company-speci c factors in which private companies di er from public companies include:

stage in life cycle;
size;
overlap of shareholders and management;
quality/depth of management;
quality of nancial and other informa on;
pressure from short-term investors; and
tax concerns.

• Stock-speci c factors that frequently a ect the value of private companies include
liquidity of equity interests in business;
concentra on of control; and
poten al agreements restric ng liquidity.

• Private company valua ons are typically performed for three di erent reasons: transac ons,
compliance ( nancial or tax repor ng), or li ga on. Acquisi on-related valua on issues and nancial
repor ng valua on issues are of greatest importance in assessing public companies.

• Di erent de ni ons (standards) of value exist. The use of a valua on and key elements pertaining to
the appraised company will help determine the appropriate de ni on. Key de ni ons of value
include
fair market value;
market value;
fair value for nancial repor ng;
fair value in a li ga on context;
investment value; and
intrinsic value.

• Private company valua ons may require adjustments to the income statement to develop es mates
of the company’s normalized earnings. Adjustments may be required for non-recurring, non-
economic, or other unusual items to eliminate anomalies and/or facilitate comparisons.

• Within the income approach, the FCF method is frequently used to value larger, mature private
companies. For smaller companies or in special situa ons, the capitalized cash ow method and
residual income method may also be used.

• Within the market approach, three methods are regularly used: the guideline public company
method, guideline transac ons method, and prior transac ons method.
ff
ti
ti
ti
fi
fi
fi
ti
fi
fi
ti
ti
ti
fi
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ff
fl
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
fi
fi
ff
ti
ti
ti
ti
fi
fl
ti
ti
ti
ti
fi
 • An asset-based approach is infrequently used in valuing private companies. This approach may be
appropriate for companies that are worth more in liquida on than as going concerns. This approach
is also applied for asset holding companies, very small companies, or companies formed recently that
have limited opera ng histories.

• Control and marketability issues are important and challenging elements in the valua on of private
companies and equity interests therein.

• If publicly traded companies are used as the basis for pricing mul ple(s), control premiums may be
appropriate in measuring the total equity value of a private company. Control premiums have also
been used to es mate lack of control discounts.

• Discounts for lack of control are used to convert a controlling interest value into a non-controlling
equity interest value. Evidence of the adverse impact of the lack of control is an important
considera on in assessing this discount.

• Discounts for lack of marketability are o en used in valuing non-controlling equity interests in private
companies. A DLOM may be inappropriate if the company has a high likelihood of a liquidity event in
the immediate future.

• Quan ca on of DLOMs can be challenging because of limited data, di erences in the interpreta on
of available data, and di erent interpreta ons of the lack of marketability’s e ect on a private
company.

• DLOM can be es mated based on 1) private sales of restricted stock in public companies rela ve to
their freely traded share price, 2) private sales of stock in companies prior to a subsequent IPO, and
3) the pricing of put op ons.
ti
fi
ti
ti
ti
ti
ti
ti
ff
ft
ti
ti
ti
ff
ff
ti
ti
ti
 Fixed Income
Learning Module 1 The Term Structure and Interest Rate Dynamics

• The spot rate for a given maturity can be expressed as a geometric average of the short-term rate and
a series of forward rates.

• Forward rates are above (below) spot rates when the spot curve is upward (downward) sloping,
whereas forward rates are equal to spot rates when the spot curve is at.

• If forward rates are realized, then all bonds, regardless of maturity, will have the same one-period
realized return, which is the rst-period spot rate.

• If the spot rate curve is upward sloping and is unchanged, then each bond “rolls down” the curve and
earns the forward rate that rolls out of its pricing (i.e., an N-period zero-coupon bond earns the N-
period forward rate as it rolls down to be a N – 1 period security). This dynamic implies an expected
return in excess of short-maturity bonds (i.e., a term premium) for longer-maturity bonds if the yield
curve is upward sloping.

• Ac ve bond por olio management is consistent with the expecta on that today’s forward curve does
not accurately re ect future spot rates.

• The swap curve provides another measure of the me value of money.

• Swaps are an essen al tool frequently used by investors to hedge, take a posi on in, or otherwise
modify interest rate risk.

• Bond quote conven ons o en use measures of spreads. Those quoted spreads can be used to
determine a bond’s price.

• Swap curves and Treasury curves can di er because of di erences in their credit exposures, liquidity,
and other supply/demand factors.

• Market par cipants o en use interest rate spreads between short-term government and risky rates
as a barometer to evaluate rela ve credit and liquidity risk.

• The local expecta ons theory, liquidity preference theory, segmented mar- kets theory, and preferred
habitat theory provide tradi onal explana ons for the shape of the yield curve.

• Historical yield curve movements suggest that they can be explained by a linear combina on of three
principal movements: level, steepness, and curvature.

• The vola lity term structure can be measured using historical data and depicts yield curve risk.
ti
ti
ti
tf
fl
ti
ti
ti
ft
ft
ti
fi
ti
ff
ti
ti
ff
ti
fl
ti
ti
 • The sensi vity of a bond value to yield curve changes may make use of e ec ve dura on, key rate
dura ons, or sensi vi es to parallel, steepness, and curvature movements. Using key rate dura ons
or sensi vi es to parallel, steepness, and curvature movements allows one to measure and manage
shaping risk.

• The term bond risk premium refers to the expected excess return of a default-free long-term bond
less that of an equivalent short-term bond or the one-period risk-free rate

• Several macroeconomic factors in uence bond pricing and required returns such as in a on,
economic growth, and monetary policy, among others.

• During highly uncertain market periods, investors ock to government bonds in a ight to quality that
is o en associated with bullish a ening, in which long-term rates fall by more than short-term rates.

• Investors expec ng rates to fall will generally extend (shorten) por olio dura on to take advantage of
expected bond price increases (decreases)

• When investors expect a steeper ( a er) curve under which long-term rates rise (fall) rela ve to
short-term rates, they will sell (buy) long-term bonds and purchase (sell) short-term bonds.
ft
ti
ti
ti
ti
ti
ti
ti
fl
fl
tt
fl
tt
fl
tf
ff
ti
ti
fl
ti
fl
ti
ti
ti
 Learning Module 2 The Arbitrage-Free Valua on Framework

This reading presents the principles and tools for arbitrage valua on of xed-income securi es. Much of
the discussion centers on the binomial interest rate tree, which can be used extensively to value both
op on-free bonds and bonds with embedded op ons. The following are the main points made in the
reading:

• A fundamental principle of valua on is that the value of any nancial asset is equal to the present
value of its expected future cash ows.

• A xed-income security is a por olio of zero-coupon bonds, each with its own discount rate that
depends on the shape of the yield curve and when the cash ow is delivered in me.

• In well-func oning markets, prices adjust un l there are no opportuni es for arbitrage, or a
transac on that involves no cash outlay yet results in a riskless pro t.

• Using the arbitrage-free approach, viewing a security as a package of zero-coupon bonds means that
two bonds with the same maturity and di erent coupon rates are viewed as di erent packages of
zero-coupon bonds and valued accordingly.

• For bonds that are op on-free, an arbitrage-free value is simply the present value of expected future
values using the benchmark spot rates.

• A binomial interest rate tree permits the short interest rate to take on one of two possible values
consistent with the vola lity assump on and an interest rate model based on a lognormal random
walk.
• An interest rate tree is a visual representa on of the possible values of interest rates (forward rates)
based on an interest rate model and an assump on about interest rate vola lity.

• The possible interest rates for any following period are consistent with the following three
assump ons: (1) an interest rate model that governs the random process of interest rates, (2) the
assumed level of interest rate vola lity, and (3) the current benchmark yield curve.

• From the lognormal distribu on, adjacent interest rates on the tree are mul ples of e raised to the 2σ
power, with the absolute change in interest rates becoming smaller and smaller as rates approach
zero.

• We use the backward induc on valua on methodology that involves start- ing at maturity, lling in
those values, and working back from right to le to nd the bond’s value at the desired node.

• The interest rate tree is t to the current yield curve by choosing interest rates that result in the
benchmark bond value. By doing this, the bond value is arbitrage free.

• An op on-free bond that is valued by using the binomial interest rate tree should have the same
value as when discoun ng by the spot rates.

• Pathwise valua on calculates the present value of a bond for each possible interest rate path and
takes the average of these values across paths.
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
tf
ti
fl
ti
ti
ti
ff
ti
ti
ti
ft
ti
ti
fi
fl
fi
ti
fi
ti
fi
ti
ti
ff
ti
fi
ti
 • The Monte Carlo method is an alterna ve method for simula ng a su ciently large number of
poten al interest rate paths in an e ort to discover how the value of a security is a ected, and it
involves randomly selec ng paths in an e ort to approximate the results of a complete pathwise
valua on.

• Term structure models seek to explain the yield curve shape and are used to value bonds (including
those with embedded op ons) and bond-related deriva ves. General equilibrium and arbitrage-free
models are the two major types of such models.

• Arbitrage-free models are frequently used to value bonds with embedded op ons. Unlike equilibrium
models, arbitrage-free models begin with the observed market prices of a reference set of nancial
instruments, and the underlying assump on is that the reference set is correctly priced.
ti
ti
ti
ti
ff
ti
ti
ff
ti
ti
ffi
ti
ff
fi
 Learning Module 3 Valua on and Analysis of Bonds with Embedded Op ons

• An embedded op on represents a right that can be exercised by the issuer, by the bondholder, or
automa cally depending on the course of interest rates. It is a ached to, or embedded in, an
underlying op on-free bond called a straight bond.

• Simple embedded op on structures include call op ons, put op ons, and extension op ons. Callable
and putable bonds can be redeemed prior to maturity, at the discre on of the issuer in the former
case and of the bond- holder in the la er case. An extendible bond gives the bondholder the right to
keep the bond for a number of years a er maturity. Putable and extendible bonds are equivalent,
except that their underlying op on-free bonds are di erent.

• Complex embedded op on structures include bonds with other types of op ons or combina ons of
op ons. For example, a conver ble bond includes a conversion op on that allows the bondholders to
convert their bonds into the issuer’s common stock. A bond with an estate put can be put by the heirs
of a deceased bondholder. Sinking fund bonds make the issuer set aside funds over me to re re the
bond issue and are o en callable, may have an accelera on provision, and may also contain a
delivery op on. Valuing and analyzing bonds with complex embedded op on structures is
challenging.

• According to the arbitrage-free framework, the value of a bond with an embedded op on is equal to
the arbitrage-free values of its parts—that is, the arbitrage-free value of the straight bond and the
arbitrage-free values of each of the embedded op ons.

• Because the call op on is an issuer op on, the value of the call op on decreases the value of the
callable bond rela ve to an otherwise iden cal but non-callable bond. In contrast, because the put
op on is an investor op on, the value of the put op on increases the value of the putable bond
rela ve to an otherwise iden cal but non-putable bond.

• In the absence of default and interest rate vola lity, the bond’s future cash ows are certain. Thus,
the value of a callable or putable bond can be calculated by discoun ng the bond’s future cash ows
at the appropriate one-period forward rates, taking into considera on the decision to exercise the
op on. If a bond is callable, the decision to exercise the op on is made by the issuer, which will
exercise the call op on when the value of the bond’s future cash ows is higher than the call price. In
contrast, if the bond is putable, the decision to exercise the op on is made by the bondholder, who
will exercise the put op on when the value of the bond’s future cash ows is lower than the put
price.

• In prac ce, interest rates uctuate and interest rate vola lity a ects the value of embedded op ons.
Thus, when valuing bonds with embedded op ons, it is important to consider the possible evolu on
of the yield curve over me.

• Interest rate vola lity is modeled using a binomial interest rate tree. The higher the vola lity, the
lower the value of the callable bond and the higher the value of the putable bond.
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ft
ti
ti
ti
ti
ti
ti
fl
ti
ti
ti
tt
ti
ft
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
tt
ti
ff
ti
fl
ti
ti
ti
ti
ti
fl
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
ti
fl
ti
 • Valuing a bond with embedded op ons assuming an interest rate vola lity requires three steps: (1)
Generate a tree of interest rates based on the given yield curve and vola lity assump ons; (2) at each
node of the tree, deter- mine whether the embedded op ons will be exercised; and (3) apply the
backward induc on valua on methodology to calculate the present value of the bond.

• The op on-adjusted spread is the single spread added uniformly to the one-period forward rates on
the tree to produce a value or price for a bond. OAS is sensi ve to interest rate vola lity: The higher
the vola lity, the lower the OAS for a callable bond.

• For bonds with embedded op ons, the best measure to assess the sensi vity of the bond’s price to a
parallel shi of the benchmark yield curve is e ec ve dura on. The e ec ve dura on of a callable or
putable bond cannot exceed that of the straight bond.

• When the op on is near the money, the convexity of a callable bond is nega ve, indica ng that the
upside for a callable bond is much smaller than the downside, whereas the convexity of a putable
bond is posi ve, indica ng that the upside for a putable bond is much larger than the downside.

• Because the prices of callable and putable bonds respond asymmetrically to upward and downward
interest rate changes of the same magnitude, one-sided dura ons provide a be er indica on
regarding the interest rate sensi vity of bonds with embedded op ons than (two-sided) e ec ve
dura on.

• Key rate dura ons show the e ect of shi ing only key points, one at a me, rather than the en re
yield curve.

• The arbitrage-free framework can be used to value capped and oored oaters. The cap provision in a
oater is an issuer op on that prevents the coupon rate from increasing above a speci ed maximum
rate. Thus, the value of a capped oater is equal to or less than the value of the straight bond. In
contrast, the oor provision in a oater is an investor op on that prevents the coupon from
decreasing below a speci ed minimum rate. Thus, the value of a oored oater is equal to or higher
than the value of the straight bond.

• The characteris cs of a conver ble bond include the conversion price, which is the applicable share
price at which the bondholders can convert their bonds into common shares, and the conversion
ra o, which re ects the number of shares of common stock that the bondholders receive from
conver ng their bonds into shares. The conversion price is adjusted in case of corporate ac ons, such
as stock splits, bonus share issuances, and rights and warrants issuances. Conver ble bondholders
may receive compensa on when the issuer pays dividends to its common shareholders, and they may
be given the opportunity to either put their bonds or convert their bonds into shares earlier and at
more advantageous terms in the case of a change of control.

• A number of investment metrics and ra os help analyze and value conver ble bonds. The conversion
value indicates the value of the bond if it is converted at the market price of the shares. The minimum
value of a conver ble bond sets a oor value for the conver ble bond at the greater of the conversion
value or the straight value. This oor is moving, however, because the straight value is not xed. The
market conversion premium represents the price investors e ec vely pay for the underlying shares if
they buy the conver ble bond and then convert it into shares. Scaled by the market price of the
shares, it represents the premium payable when buying the conver ble bond rather than the
underlying common stock.
fl
ti
ti
ti
ti
ti
ft
ti
ti
ti
fl
fl
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ff
ti
ti
fl
fl
fl
fl
ti
ti
ft
ff
ti
ti
ti
ti
ti
ti
ff
ti
fl
ti
fl
ti
ti
ff
ti
ti
fl
ti
ti
fl
ti
ti
ti
tt
ti
ti
ti
ti
fi
ti
ti
ff
fi
ti
ti
ti
 • Because conver ble bonds combine characteris cs of bonds, stocks, and op ons, as well as
poten ally other features, their valua on and analysis are challenging. Conver ble bond investors
should consider the factors that a ect not only bond prices but also the underlying share price.

• The arbitrage-free framework can be used to value conver ble bonds, including callable and putable
ones. Each component (straight bond, call op on of the stock, and call and/or put op on on the
bond) can be valued separately.

• The risk–return characteris cs of a conver ble bond depend on the under- lying share price rela ve
to the conversion price. When the underlying share price is well below the conversion price, the
conver ble bond is “busted” and exhibits mostly bond risk–return characteris cs. Thus, it is mainly
sensi ve to interest rate movements. In contrast, when the underlying share price is well above the
conversion price, the conver ble bond exhibits mostly stock risk–return characteris cs. Thus, its price
follows similar movements to the price of the underlying stock. In between these two extremes, the
conver ble bond trades like a hybrid instrument.
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
 Learning Module 4 Credit Analysis Models

We have covered several important topics in credit analysis. Among the points made are the following:

• Three factors important to modeling credit risk are the expected exposure to default, the recovery
rate, and the loss given default.

• These factors permit the calcula on of a credit valua on adjustment that is subtracted from the
(hypothe cal) value of the bond, if it were default risk free, to get the bond’s fair value given its credit
risk. The credit valua on adjustment is calculated as the sum of the present values of the expected
loss for each period in the remaining life of the bond. Expected values are computed using risk-
neutral probabili es, and discoun ng is done at the risk-free rates for the relevant maturi es.

• The CVA captures investors’ compensa on for bearing default risk. The compensa on can also be
expressed in terms of a credit spread.

• Credit scores and credit ra ngs are third-party evalua ons of creditworthiness used in dis nct
markets.

• Analysts may use credit ra ngs and a transi on matrix of probabili es to adjust a bond’s yield to
maturity to re ect the probabili es of credit migra on. Credit spread migra on typically reduces
expected return.

• Credit analysis models fall into two broad categories: structural models and reduced-form models.

• Structural models are based on an op on perspec ve of the posi ons of the stakeholders of the
company. Bondholders are viewed as owning the assets of the company; shareholders have call
op ons on those assets.

• Reduced-form models seek to predict when a default may occur, but they do not explain the why as
structural models do. Reduced-form models, unlike structural models, are based only on observable
variables.

• When interest rates are assumed to be vola le, the credit risk of a bond can be es mated in an
arbitrage-free valua on framework.

• The discount margin for oa ng-rate notes is similar to the credit spread for xed-coupon bonds. The
discount margin can also be calculated using an arbitrage-free valua on framework.

• Arbitrage-free valua on can be applied to judge the sensi vity of the credit spread to changes in
credit risk parameters.

• The term structure of credit spreads depends on macro and micro factors.

• As it concerns macro factors, the credit spread curve tends to become steeper and to widen in
condi ons of weak economic ac vity. Market supply and demand dynamics are important. The most
frequently traded securi es tend to determine the shape of this curve.

• Issuer- or industry-speci c factors, such as the chance of a future leverage-decreasing event, can
cause the credit spread curve to a en or invert.
ti
ti
ti
fl
ti
ti
ti
ti
fi
ti
fl
ti
ti
ti
ti
ti
ti
fl
ti
tt
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
 • When a bond is very likely to default, it o en trades close to its recovery value at various maturi es;
moreover, the credit spread curve is less informa ve about the rela onship between credit risk and
maturity.

• For securi zed debt, the characteris cs of the asset por olio themselves suggest the best approach
for a credit analyst to take when deciding among investments. Important considera ons include the
rela ve concentra on of assets and their similarity or heterogeneity as it concerns credit risk.
ti
ti
ti
ti
ft
ti
tf
ti
ti
ti
 Learning Module 5 Credit Default Swaps

• A credit default swap (CDS) is a contract between two par es in which one party purchases
protec on from another party against losses from the default of a borrower for a de ned period of
me.

• A CDS is wri en on the debt of a third party, called the reference en ty, whose relevant debt is called
the reference obliga on, typically a senior unsecured bond.

• A CDS wri en on a par cular reference obliga on normally provides coverage for all obliga ons of
the reference en ty that have equal or higher seniority.

• The two par es to the CDS are the credit protec on buyer, who is said to be short the reference
en ty’s credit, and the credit protec on seller, who is said to be long the reference en ty’s credit.

• The CDS pays o upon occurrence of a credit event, which includes bankruptcy, failure to pay, and, in
some countries, involuntary restructuring.

• Se lement of a CDS can occur through a cash payment from the credit protec on seller to the credit
protec on buyer as determined by the cheapest-to-deliver obliga on of the reference en ty or by
physical delivery of the reference obliga on from the protec on buyer to the protec on seller in
exchange for the CDS no onal.

• A cash se lement payo is determined by an auc on of the reference en ty’s debt, which gives the
market’s assessment of the likely recovery rate. The credit protec on buyer must accept the outcome
of the auc on even though the ul mate recovery rate could di er.

• CDS can be constructed on a single en ty or as indexes containing mul ple en es. Bespoke CDS or
baskets of CDS are also common.

• The xed payments made from CDS buyer to CDS seller are customarily set at a xed annual rate of
1% for investment-grade debt or 5% for high-yield debt.

• Valua on of a CDS is determined by es ma ng the present value of the payment leg, which is the
series of payments made from the protec on buyer to the protec on seller, and the present value of
the protec on leg, which is the payment from the protec on seller to the protec on buyer in event
of default. If the present value of the payment leg is greater than the present value of the protec on
leg, the protec on buyer pays an upfront premium to the seller. If the present value of the protec on
leg is greater than the present value of the payment leg, the seller pays an upfront premium to the
buyer.

• An important determinant of the value of the expected payments is the hazard rate, the probability
of default given that default has not already occurred.

• CDS prices are o en quoted in terms of credit spreads, the implied number of basis points that the
credit protec on seller receives from the credit protec on buyer to jus fy providing the protec on.

• Credit spreads are o en expressed in terms of a credit curve, which expresses the rela onship
between the credit spreads on bonds of di erent maturi es for the same borrower.
ti
tt
ti
fi
ti
ti
ti
tt
tt
ti
ti
tt
ti
ti
ti
ff
ft
ti
ft
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
 • CDS change in value over their lives as the credit quality of the reference en ty changes, which leads
to gains and losses for the counterpar es, even though default may not have occurred or may never
occur. CDS spreads approach zero as the CDS approaches maturity.

• Either party can mone ze an accumulated gain or loss by entering into an o se ng posi on that
matches the terms of the original CDS.

• CDS are used to increase or decrease credit exposures or to capitalize on di erent assessments of the
cost of credit among di erent instruments ed to the reference en ty, such as debt, equity, and
deriva ves of debt and equity.
ti
ti
ff
ti
ti
ti
ff
ff
ti
tti
ti
 Derivatives
Learning Module 1 Pricing and Valua on of Forward Commitments

This reading on forward commitment pricing and valua on provides a founda on for understanding how
forwards, futures, and swaps are both priced and valued.

Key points include the following:

• The arbitrageur would rather have more money than less and abides by two fundamental rules: Do
not use your own money, and do not take any price risk.

• The no-arbitrage approach is used for the pricing and valua on of forward commitments and is built
on the key concept of the law of one price, which states that if two investments have the same future
cash ows, regardless of what happens in the future, these two investments should have the same
current price.

• Throughout this reading, the following key assump ons are made:
Replica ng and o se ng instruments are iden able and investable.
Market fric ons are nil.
Short selling is allowed with full use of proceeds.
Borrowing and lending are available at a known risk-free rate.

• Carry arbitrage models used for forward commitment pricing and valua on are based on the no-
arbitrage approach.

• With forward commitments, there is a dis nct di erence between pricing and valua on. Pricing
involves the determina on of the appropriate xed price or rate, and valua on involves the
determina on of the contract’s cur- rent value expressed in currency units.

• Forward commitment pricing results in determining a price or rate such that the forward contract
value is equal to zero.

• Using the carry arbitrage model, the forward contract price (F0) is:

• The key forward commitment pricing equa ons with carry costs (CC) and carry bene ts (CB) are:
fl
ti
ti
ti
ff
tti
ti
ti
ti
ti
fi
ti
fi
ff
ti
ti
ti
ti
ti
ti
ti
fi
 • Futures contract pricing in this reading can essen ally be treated the same as for- ward contract
pricing.

• The value of a forward commitment is a func on of the price of the underly- ing instrument, nancing
costs, and other carry costs and bene ts.

• The key forward commitment valua on equa ons are :

• With equi es and xed-income securi es, the forward price is determined such that the ini al
forward value is zero.

• A forward rate agreement (FRA) is a forward contract on interest rates. The FRA’s xed interest rate is
determined such that the ini al value of the FRA is zero.

• FRA se lements amounts at Time h are:

• The FRA’s xed interest rate (annualized) at contract ini a on is:

• The Time g value of an FRA ini ated at Time 0 is:

tt
ti
fi
fi
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
fi
ti
fi
 • The xed-income forward (or futures) price including conversion factor (i.e., adjusted price) is:

and the conversion factor adjusted futures price (i.e., quoted futures price) is:

• The general approach to pricing and valuing swaps as covered here is using a replica ng por olio or
o se ng por olio of comparable instruments, typically bonds for interest rate and currency swaps
and equi es plus bonds for equity swaps.

• The swap pricing equa on, which sets rFIX for the implied xed
bond in an interest rate swap, is:

• The value of an interest rate swap at a point in Time t a er ini a on is the sum of the present values
of the di erence in xed swap rates mes the stated no onal amount, or:

• With a basic understanding of pricing and valuing a simple interest rate swap, it is a straigh orward
extension to pricing and valuing currency swaps and equity swaps.

• The solu on for each of the three variables, one no onal amount (NAa) and two xed rates (one for
each currency, a and b), needed to price a xed-for- xed currency swap are :

• The currency swap valua on equa on, for valuing the swap at me t (a er ini a on), can be
expressed as:
ff
fi
tti
ff
ti
ti
tf
fi
ti
ti
ti
ti
fi
ti
fi
ft
ti
fi
ti
ti
ti
ft
ti
ti
fi
ti
tf
tf
 • For a receive- xed, pay equity swap, the xed rate (rFIX) for the implied xed bond that makes the
swap value equal to “0” at ini a on is:

• The value of an equity swap at Time t (VEQ,t), a er ini a on, is:

where VFIX (C0) is the Time t value of a xed-rate bond ini ated with coupon C0 at Time 0, St is the
current equity price, St–1 is the equity price at the last reset date, and PV() is the PV func on from the
swap maturity date to Time t.
fi
ti
ti
fi
fi
ft
ti
ti
ti
fi
ti
 Learning Module 2 Valua on of Con ngent Claims

This reading on the valua on of con ngent claims provides a founda on for under- standing how a
variety of di erent op ons are valued. Key points include the following:

• The arbitrageur would rather have more money than less and abides by two fundamental rules: Do
not use your own money and do not take any price risk.

• The no-arbitrage approach is used for op on valua on and is built on the key concept of the law of
one price, which says that if two investments have the same future cash ows regardless of what
happens in the future, then these two investments should have the same current price.

• Throughout this reading, the following key assump ons are made:
Replica ng instruments are iden able and investable.
Market fric ons are nil.
Short selling is allowed with full use of proceeds.
The underlying instrument price follows a known distribu on.
Borrowing and lending is available at a known risk-free rate.

• The two-period binomial model can be viewed as three one-period binomial models, one posi oned
at Time 0 and two posi oned at Time 1.

• In general, European-style op ons can be valued based on the expecta ons approach in which the
op on value is determined as the present value of the expected future op on payouts, where the
discount rate is the risk-free rate and the expecta on is taken based on the risk-neutral probability
measure.

• Both American-style op ons and European-style op ons can be valued based on the no-arbitrage
approach, which provides clear interpreta ons of the component terms; the op on value is
determined by working backward through the binomial tree to arrive at the correct current value.

• For American-style op ons, early exercise in uences the op on values and hedge ra os as one works
backward through the binomial tree.

• Interest rate op on valua on requires the speci ca on of an en re term structure of interest rates,
so valua on is o en es mated via a binomial tree.

• A key assump on of the Black–Scholes–Merton op on valua on model is that the return of the
underlying instrument follows geometric Brownian mo on, implying a lognormal distribu on of the
price.

• The BSM model can be interpreted as a dynamically managed por olio of the underlying instrument
and zero-coupon bonds.

• BSM model interpreta ons related to N(d1) are that it is the basis for the number of units of
underlying instrument to replicate an op on, that it is the primary determinant of delta, and that it
answers the ques on of how much the op on value will change for a small change in the underlying.
ti
ti
ti
ti
ff
ti
ti
ft
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
fl
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
tf
ti
ti
fl
ti
ti
ti
ti
ti
 • BSM model interpreta ons related to N(d2) are that it is the basis for the number of zero-coupon
bonds to acquire to replicate an op on and that it is the basis for es ma ng the risk-neutral
probability of an op on expiring in the money.

• The Black futures op on model assumes the underlying is a futures or a forward contract.

• Interest rate op ons can be valued based on a modi ed Black futures op on model in which the
underlying is a forward rate agreement (FRA), there is an accrual period adjustment as well as an
underlying no onal amount, and that care must be given to day-count conven ons.

• An interest rate cap is a por olio of interest rate call op ons termed caplets, each with the same
exercise rate and with sequen al maturi es.

• An interest rate oor is a por olio of interest rate put op ons termed oor- lets, each with the same
exercise rate and with sequen al maturi es.

• A swap on is an op on on a swap.

• A payer swap on is an op on on a swap to pay xed and receive oa ng.

• A receiver swap on is an op on on a swap to receive xed and pay oa ng.

• Long a callable xed-rate bond can be viewed as long a straight xed-rate bond and short a receiver
swap on.

• Delta is a sta c risk measure de ned as the change in a given por olio for a given small change in the
value of the underlying instrument, holding every- thing else constant.

• Delta hedging refers to managing the por olio delta by entering addi onal posi ons into the
por olio.

• A delta neutral por olio is one in which the por olio delta is set and maintained at zero.

• A change in the op on price can be es mated with a delta approxima on.

• Because delta is used to make a linear approxima on of the non-linear rela onship that exists
between the op on price and the underlying price, there is an error that can be es mated by gamma.

• Gamma is a sta c risk measure de ned as the change in a given por olio delta for a given small
change in the value of the underlying instrument, holding everything else constant.

• Gamma captures the non-linearity risk or the risk—via exposure to the underlying—that remains
once the por olio is delta neutral.

• A gamma neutral por olio is one in which the por olio gamma is maintained at zero.

• The change in the op on price can be be er es mated by a delta-plus-gamma approxima on

compared with just a delta approxima on.
tf
ti
ti
ti
tf
ti
ti
ti
fi
ti
ti
ti
fl
ti
tf
ti
ti
ti
ti
tf
ti
ti
tf
ti
tf
ti
ti
fi
fi
ti
ti
ti
ti
ti
tt
tf
ti
tf
fi
ti
tf
fi
fi
ti
ti
fi
tf
fl
fl
ti
tf
ti
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
ti
 • Theta is a sta c risk measure de ned as the change in the value of an op on given a small change in
calendar me, holding everything else constant.

• Vega is a sta c risk measure de ned as the change in a given por olio for a given small change in
vola lity, holding everything else constant.

• Rho is a sta c risk measure de ned as the change in a given por olio for a given small change in the
risk-free interest rate, holding everything else constant.

• Although historical vola lity can be es mated, there is no objec ve measure of future vola lity.

• Implied vola lity is the BSM model vola lity that yields the market op on price.

• Implied vola lity is a measure of future vola lity, whereas historical vola lity is a measure of past
vola lity.

• Op on prices re ect the beliefs of op on market par cipant about the future vola lity of the
underlying.

• The vola lity smile is a two dimensional plot of the implied vola lity with respect to the exercise
price.

• The vola lity surface is a three dimensional plot of the implied vola lity with respect to both
expira on me and exercise prices.

• If the BSM model assump ons were true, then one would expect to nd the vola lity surface at, but
in prac ce, the vola lity surface is not at.
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fl
ti
ti
ti
fi
fi
fi
ti
ti
fl
ti
ti
ti
ti
ti
tf
tf
ti
fi
ti
ti
ti
ti
ti
ti
fl
 Alternative Investments
Learning Module 1 Overview of Types of Real Estate Investment

General Characteris cs of Real Estate

• Real estate investments can occur in four basic forms: private equity (direct ownership), publicly
traded equity (indirect ownership claim), private debt (direct mortgage lending), and publicly traded
debt (securi zed mortgages).

• Many mo va ons exist for inves ng in real estate income property. The key ones are current income,
price apprecia on, in a on hedge, diversi ca on, and tax bene ts.

• Adding equity real estate investments to a tradi onal por olio will poten ally have diversi ca on
bene ts because of the less-than-perfect correla on of equity real estate returns with returns to
stocks and bonds.

• If the income stream can be adjusted for in a on and real estate prices increase with in a on, then
equity real estate investments may provide an in a on hedge.

• Debt investors in real estate expect to receive their return from promised cash ows and typically do
not par cipate in any apprecia on in value of the underlying real estate. Thus, debt investments in
real estate are similar to other xed-income investments, such as bonds.

• Regardless of the form of real estate investment, the value of the underlying real estate property can
a ect the performance of the investment with loca on being a cri cal factor in determining the value
of a real estate property.

• Real estate property has some unique characteris cs compared with other investment asset classes.
These characteris cs include heterogeneity and xed loca on, high unit value, management
intensiveness, high transac on costs, deprecia on, sensi vity to the credit market, illiquidity, and
di culty of value and price determina on.

• There are many di erent types of real estate proper es in which to invest. The main commercial
(income-producing) real estate property types are o ce, industrial and warehouse, retail, and mul -
family. Other types of commercial proper es are typically classi ed by their speci c use.

• Certain risk factors are common to commercial property, but each property type is likely to have a
di erent suscep bility to these factors. The key risk factors that can a ect commercial real estate
include business condi ons, lead me for new development, excess supply, cost and availability of
capital, unexpected in a on, demographics, lack of liquidity, environmental issues, availability of
informa on, management exper se, and leverage.

• Loca on, lease structures, and economic factors—such as economic growth, popula on growth,
employment growth, and consumer spending—a ect the value of each property type.
ff
ffi
ff
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
fl
fl
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
fi
fl
ti
ti
ti
ti
fi
ti
fl
ff
ti
ti
ti
ffi
ti
ti
tf
ti
fi
fi
ti
ff
ti
fl
fi
ti
fl
fi
ti
ti
ti
 Learning Module 2 Investments in Real Estate through Private Vehicles

• Generally, three di erent valua on approaches are used by appraisers: income, cost, and sales
comparison.

• The income approach includes direct capitaliza on and discounted cash ow methods. Both methods
focus on net opera ng income as an input to the value of a property and indirectly or directly factor
in expected growth.

• The cost approach es mates the value of a property based on adjusted replacement cost. This
approach is typically used for unusual proper es for which market comparables are di cult to obtain.

• The sales comparison approach es mates the value of a property based on what price comparable
proper es are selling for in the current market.

• When debt nancing is used to purchase a property, addi onal ra os and returns calculated and
interpreted by debt and equity investors include the loan-to-value ra o, the debt service coverage
ra o, and leveraged and unleveraged internal rates of return.
ti
ti
fi
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
fl
ffi
 Learning Module 3 Investments in Real Estate Through Publicly Traded Securi es

• The principal types of publicly traded real estate securi es include real estate investment trusts
(REITs), real estate opera ng companies (REOCs), and residen al and commercial mortgage-backed
securi es (RMBS and CMBS).

• Compared with other publicly traded shares, REITs typically o er higher-than-average yields and
greater stability of income and returns. They are amenable to a net asset value approach to valua on
because of the existence of ac ve private markets for their real estate assets.

• Compared with REOCs, REITs o er higher yields and income tax exemp ons but have less opera ng
exibility to invest in a broad range of real estate ac vi es and less poten al for growth from
reinves ng their opera ng cash ows because of their high income-to-payout ra os.

• In assessing the investment merits of REITs, investors analyze the e ects of trends in general
economic ac vity, retail sales, job crea on, popula on growth, and new supply and demand for
speci c types of space. Investors also pay par cular a en on to occupancies, leasing ac vity, rental
rates, remaining lease terms, in-place rents compared with market rents, costs to maintain space and
re-lease space, tenants’ nancial health and tenant concentra on in the por olio, nancial leverage,
debt maturi es and costs, and the quality of management and governance.

• Analysts make adjustments to the historical cost-based nancial statements of REITs and REOCs to
obtain be er measures of current income and net worth. The three principal gures they calculate
and use are (1) funds from opera ons or accoun ng net earnings, excluding deprecia on, deferred
tax charges, and gains or losses on sales of property and debt restructuring; (2) adjusted funds from
opera ons, or funds from opera ons adjusted to remove straight-line rent and to provide for
maintenance-type capital expenditures and leasing costs, including leasing agents’ commissions and
tenants’ improvement allowances; and (3) net asset value or the di erence between a real estate
company’s asset and liability ranking prior to share- holders’ equity, all valued at market values
instead of accoun ng book values.

• REITs and some REOCs generally return a signi cant por on of their income to their investors as
required by law and, as a result, tend to pay high dividends. Thus, dividend discount or discounted
cash ow models for valua on are also applicable. These valua on approaches are applied in the
same manner as they are for shares in other industries. Usually, investors use two- or three-step
dividend discount models with near-term, intermediate-term, and/or long-term growth assump ons.
In discounted cash ow models, investors o en use intermediate-term cash ow projec ons and a
terminal value based on historical cash ow mul ples.
fl
fl
fi
ti
ti
ti
tt
ti
ti
ti
fl
ti
fi
ti
ti
ti
ff
ti
fl
ti
ti
fl
ft
ti
fi
ti
ti
ti
ti
tt
ti
ti
fi
ti
ti
ff
ti
ti
ti
ff
ff
ti
ti
tf
fl
fi
ti
ti
fi
ti
ti
ti
ti
ti
ti
 Learning Module 4 Private Equity Investments

• Private equity funds seek to add value by various means, including op mizing nancial structures,
incen vizing management, and crea ng opera onal improvements.

• Private equity can be thought of as an alterna ve system of governance for corpora ons: Rather than
ownership and control being separated as in most publicly quoted companies, private equity
concentrates ownership and control. Many view the combina on of ownership and control as a
fundamental source of the returns earned by the best private equity funds.

• A cri cal role for the GP is valua on of poten al investments. But because these investments are
usually privately owned, valua on encounters many challenges.

• Valua on techniques di er according to the nature of the investment. Early-stage ventures require
very di erent techniques than leveraged buyouts. Private equity professionals tend to use mul ple
techniques when performing a valua on, and they explore many di erent scenarios for the future
development of the business.

• In buyouts, the availability of debt nancing can have a big impact on the scale of private equity
ac vity, and it seems to impact valua ons observed in the market.

• Because private equity funds are incen vized to acquire, add value, and then exit within the life me
of the fund, they are considered buy-to-sell investors. Planning the exit route for the investment is a
cri cal role for the GP, and a well- med and well-executed investment can be a signi cant source of
realized value.

• In addi on to the problems encountered by the private equity funds in valuing poten al por olio
investments, challenges exist in valuing the investment por olio on an ongoing basis. This is because
the investments have no easily observed market value and there is a large element of judgment
involved in valuing each of the por olio companies prior to their sale by the fund.

• The two main metrics for measuring the ongoing and ul mate performance of private equity funds
are IRR and mul ples. Comparisons of PE returns across funds and with other assets are demanding
because it is important to control for the ming of cash ows, di erences in risk and por olio
composi on, and vintage-year e ects.
ti
ti
ti
ti
ti
ff
ti
ti
ti
ff
ti
ti
ff
ti
tf
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
fl
tf
ti
ff
ff
ti
fi
ti
fi
ti
tf
tf
ti
ti
 Learning Module 5 Introduc on to Commodi es and Commodity Deriva ves

• Commodi es are a diverse asset class comprising various sectors: energy, grains, industrial (base)
metals, livestock, precious metals, and so s (cash crops). Each of these sectors has a number of
characteris cs that are important in determining the supply and demand for each commodity,
including ease of storage, geopoli cs, and weather.

• Fundamental analysis of commodi es relies on analyzing supply and demand for each of the products
as well as es ma ng the reac on to the inevitable shocks to their equilibrium or underlying direc on.

• The life cycle of commodi es varies considerably depending on the economic, technical, and
structural (i.e., industry, value chain) pro le of each commodity as well as the sector. A short life cycle
allows for rela vely rapid adjustment to outside events, whereas a long life cycle generally limits the
ability of the market to react.

• The valua on of commodi es rela ve to that of equi es and bonds can be summarized by no ng
that equi es and bonds represent nancial assets whereas commodi es are physical assets. The
valua on of commodi es is not based on the es ma on of future pro tability and cash ows but
rather on a discounted forecast of future possible prices based on such factors as the supply and
demand of the physical item.

• The commodity trading environment is similar to other asset classes, with three types of trading
par cipants: (1) informed investors/hedgers, (2) speculators, and (3) arbitrageurs.

• Commodi es have two general pricing forms: spot prices in the physical markets and futures prices
for later delivery. The spot price is the current price to deliver or purchase a physical commodity at a
speci c loca on. A futures price is an exchange-based price agreed on to deliver or receive a de ned
quan ty and o en quality of a commodity at a future date.

• The di erence between spot and futures prices is generally called the basis. When the spot price is
higher than the futures price, it is called backwarda on, and when it is lower, it is called contango.
Backwarda on and contango are also used to describe the rela onship between two futures
contracts of the same commodity.

• Commodity contracts can be se led by either cash or physical delivery.

• There are three primary theories of futures returns.

In insurance theory, commodity producers who are long the physical good are mo ved to sell the
commodity for future delivery to hedge their produc on price risk exposure.
The hedging pressure hypothesis describes when producers along with consumers seek to protect
themselves from commodity market price vola lity by entering into price hedges to stabilize their
projected pro ts and cash ow.
The theory of storage focuses on supply and demand dynamics of commodity inventories,
including the concept of “convenience yield.”

• The total return of a fully collateralized commodity futures contract can be quan ed as the spot
price return plus the roll return plus the collateral return (risk-free rate return).
ti
ti
fi
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ft
ti
ti
ti
ti
fl
ti
ti
tt
ti
ti
ti
fi
fi
ft
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
fi
ti
fl
ti
fi
ti
 • The roll return is e ec vely the weighted accoun ng di erence (in percent- age terms) between the
near-term commodity futures contract price and the farther-term commodity futures contract price.

• A commodity swap is a legal contract between two par es calling for the exchange of payments over
mul ple dates as determined by several reference prices or indexes.

• The most relevant commodity swaps include excess return swaps, total return swaps, basis swaps,
and variance/vola lity swaps.

• The ve primary commodity indexes based on assets are (1) the S&P GSCI; (2) the Bloomberg
Commodity Index, formerly the Dow Jones–UBS Commodity Index; (3) the Deutsche Bank Liquid
Commodity Index; (4) the Thomson Reuters/CoreCommodity CRB Index; and (5) the Rogers
Interna onal Commodi es Index.

• The key di eren a ng characteris cs of commodity indexes are

the breadth and selec on methodology of coverage (number of commodi es and sectors)
included in each index, no ng that some commodi es have mul ple reference contracts,
the rela ve weigh ngs assigned to each component/commodity and the related methodology for
how these weights are determined,
the methodology and frequency for rolling the individual futures contracts,
the methodology and frequency for rebalancing the weights of the individual commodi es and
sectors, and
the governance that determines which commodi es are selected.
ti
fi
ti
ti
ff
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
 Portfolio Management
Learning Module 1 Exchange-Traded Funds: Mechanics and Applica ons

We have examined important considera ons for ETF investors, including how ETFs work and trade, tax
e cient a ributes, and key por olio uses. The following is a summary of key points:

• ETFs rely on a crea on/redemp on mechanism that allows for the con nuous crea on and
redemp on of ETF shares.

• The only investors who can create or redeem new ETF shares are a special group of ins tu onal
investors called authorized par cipants.

• ETFs trade on both the primary market (directly between APs and issuers) and on the secondary
markets (exchange-based or OTC trades, such as listed equity).

• End investors trade ETFs on the secondary markets, like stocks.

• Holding period performance devia ons (tracking di erences) are more useful than the standard
devia on of daily return di erences (tracking error).

• ETF tracking di erences from the index occur for the following reasons:
fees and expenses,
representa ve sampling/op miza on,
use of depositary receipts and other ETFs,
index changes,
fund accoun ng prac ces,
regulatory and tax requirements, and
asset manager opera ons.

• ETFs are generally taxed in the same manner as the securi es they hold, with some nuances:
ETFs are more tax fair than tradi onal mutual funds, because por olio trading is generally not
required when money enters or exits an ETF.
Owing to the crea on/redemp on process, ETFs can be more tax e cient than mutual funds.
ETF issuers can redeem out low-cost-basis securi es to minimize future taxable gains.
Local markets have unique ETF taxa on issues that should be considered.

• ETF bid–ask spreads vary by trade size and are usually published for smaller trade sizes. They are
ghtest for ETFs that are very liquid and have con nuous two-way order ow. For less liquid ETFs, the
following factors can determine the quoted bid–ask spread of an ETF trade:
Crea on/redemp on costs, brokerage and exchange fees
Bid–ask spread of underlying securi es held by the ETF
Risk of hedging or carry posi ons by liquidity provider
Market makers’ target pro t spread
ti
ffi
ti
ti
ti
tt
ti
ti
ff
ti
ti
ti
ti
ti
fi
ff
ti
ti
ti
ti
ti
tf
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
tf
ffi
ti
ti
fl
ti
ti
ti
 • ETF bid–ask spreads on xed income rela ve to equity tend to be wider because the underlying
bonds trade in dealer markets and hedging is more di cult. Spreads on ETFs holding interna onal
stocks are ghtest when the underlying security markets are open for trading.

• ETF premiums and discounts refer to the di erence between the exchange price of the ETF and the
fund’s calculated NAV, based on the prices of the underlying securi es and weighted by the por olio
posi ons at the start

• of each trading day. Premiums and discounts can occur because NAVs are based on the last traded
prices, which may be observed at a me lag to the ETF price, or because the ETF is more liquid and
more re ec ve of current informa on and supply and demand than the underlying securi es in
rapidly changing markets.

• Costs of ETF ownership may be posi ve or nega ve and include both explicit and implicit costs. The
main components of ETF cost are
the fund management fee;
tracking error;
por olio turnover;
trading costs, such as commissions, bid–ask spreads, and premiums/ discounts;
taxable gains/losses; and
security lending.

• Trading costs are incurred when the posi on is entered and exited. These one- me costs decrease as
a por on of total holding costs over longer holding periods and are a more signi cant considera on
for shorter-term tac cal ETF traders.

• Other costs, such as management fees and por olio turnover, increase as a propor on of overall cost
as the investor holding period lengthens. These costs are a more signi cant considera on for longer-
term buy-and-hold investors.

• ETFs are di erent from exchange-traded notes, although both use the crea on/redemp on process.
Exchange-traded notes carry unique counterparty risks of default.
Swap-based ETFs may carry counterparty risk.
ETFs, like mutual funds, may lend their securi es, crea ng risk of counterparty default.
ETF closures can create unexpected tax liabili es.

• ETFs are used for core asset class exposure, mul -asset, dynamic, and tac cal strategies based on
investment views or changing market condi ons; for factor or smart beta strategies with a goal to
improve return or modify por olio risk; and for por olio e ciency applica ons, such as rebalancing,
liquidity management, comple on strategies, and transi ons.

• ETFs are useful for inves ng cash in ows, as well as for raising proceeds to provide for client
withdrawals. ETFs are used for rebalancing to target asset class weights and for “comple on
strategies” to ll a temporary gap in an asset class category, sector, or investment theme or when
external managers are underweight. When posi ons are in transi on from one external man- ager to
another, ETFs are o en used as the temporary holding and may be used to fund the new manager.

• All types of investors use ETFs to establish low-cost core exposure to asset classes, equity style
benchmarks, xed-income categories, and commodi es.
tf
ti
ti
fl
ti
ff
ti
fi
fi
ft
ti
fi
ti
tf
ti
ti
fl
ti
ti
ti
ti
ff
ti
ti
ti
tf
ti
ti
ti
tf
ti
ffi
ti
ti
ffi
ti
ti
fi
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
tf
ti
 • For more tac cal inves ng, thema c ETFs are used in ac ve por olio management and represent
narrow or niche areas of the equity market not well represented by industry or sector ETFs.

• Systema c, ac ve strategies that use rules-based benchmarks for exposure to such factors as size,
value, momentum, quality, or dividend lts or combina ons of these factors are frequently
implemented with ETFs.
• Mul -asset and global asset alloca on or macro strategies that manage posi ons dynamically as
market condi ons change are also areas where ETFs are frequently used.

• Proper u liza on requires investors to carefully research and assess the ETF’s index construc on
methodology, costs, risks, and performance history.
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
tf
ti
ti
 Learning Module 2 Using Mul factor Models

In our coverage of mul factor models, we have presented concepts, models, and tools that are key
ingredients to quan ta ve por olio management and are used to both construct por olios and to
a ribute sources of risk and return.

• Mul factor models permit a nuanced view of risk that is more granular than the single-factor
approach allows.

• Mul factor models describe the return on an asset in terms of the risk of the asset with respect to a
set of factors. Such models generally include systema c factors, which explain the average returns of
a large number of risky assets. Such factors represent priced risk—risk for which investors require an
addi onal return for bearing.

• The arbitrage pricing theory (APT) describes the expected return on an asset (or por olio) as a linear
func on of the risk of the asset with respect to a set of factors. Like the CAPM, the APT describes a
nancial market equilibrium; however, the APT makes less strong assump ons.

• The major assump ons of the APT are as follows:

Asset returns are described by a factor model.
With many assets to choose from, asset-speci c risk can be eliminated.
Assets are priced such that there are no arbitrage opportuni es.

• Mul factor models are broadly categorized according to the type of factor used:
Macroeconomic factor models
Fundamental factor models
Sta s cal factor models

• In macroeconomic factor models, the factors are surprises in macroeconomic variables that
signi cantly explain asset class (equity in our examples) returns. Surprise is de ned as actual minus
forecasted value and has an expected value of zero. The factors can be understood as a ec ng either
the expected future cash ows of companies or the interest rate used to dis- count these cash ows
back to the present and are meant to be uncorrelated.

• In fundamental factor models, the factors are a ributes of stocks or companies that are important in
explaining cross-sec onal di erences in stock prices. Among the fundamental factors are book-value-
to-price ra o, mar- ket capitaliza on, price-to-earnings ra o, and nancial leverage.

• In contrast to macroeconomic factor models, in fundamental models the factors are calculated as
returns rather than surprises. In fundamental factor models, we generally specify the factor
sensi vi es (a ributes) rst and then es mate the factor returns through regressions. In
macroeconomic factor models, however, we rst develop the factor (surprise) series and then
es mate the factor sensi vi es through regressions. The factors of most fundamental factor models
may be classi ed as company fundamental factors, company share-related factors, or macroeconomic
factors.

• In sta s cal factor models, sta s cal methods are applied to a set of historical returns to determine
por olios that explain historical returns in one of two senses. In factor analysis models, the factors
are the por olios that best explain (reproduce) historical return covariances. In principal-components
models, the factors are por olios that best explain (reproduce) the historical return variances.
fi
tt
ti
ti
tf
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
tf
fi
tt
ti
ti
ti
ti
ti
fi
ti
fl
tf
ti
ff
ti
ti
tf
ti
ti
ti
fi
fi
tt
ti
ti
ti
fi
ti
fi
tf
tf
ff
ti
fl
 • Mul factor models have applica ons to return a ribu on, risk a ribu on, por olio construc on, and
strategic investment decisions.

• A factor por olio is a por olio with unit sensi vity to a factor and zero sensi vity to other factors.

• Ac ve return is the return in excess of the return on the benchmark.

• Ac ve risk is the standard devia on of ac ve returns. Ac ve risk is also called tracking error or
tracking risk. Ac ve risk squared can be decomposed as the sum of ac ve factor risk and ac ve
speci c risk.

• The informa on ra o (IR) is mean ac ve return divided by ac ve risk (tracking error). The IR measures
the increment in mean ac ve return per unit of ac ve risk.

• Factor models have uses in construc ng por olios that track market indexes and in alterna ve index
construc on.

• Tradi onally, the CAPM approach would allocate assets between the risk-free asset and a broadly
diversi ed index fund. Considering mul ple sources of systema c risk may allow investors to improve
on that result by l ng away from the market por olio. Generally, investors would gain from
accep ng above average (below average) exposures to risks that they have a compara ve advantage
(compara ve disadvantage) in bearing.
ti
ti
ti
ti
fi
ti
fi
ti
ti
tf
ti
ti
ti
ti
ti
tf
ti
ti
ti
ti
ti
ti
ti
tf
ti
tt
tf
ti
ti
ti
ti
ti
tt
ti
ti
ti
tf
ti
ti
ti
ti
 Learning Module 3 Measuring and Managing Market Risk

This reading on market risk management models covers various techniques used to manage the risk
arising from market uctua ons in prices and rates. The key points are summarized as follows:

• Value at risk (VaR) is the minimum loss in either currency units or as a percentage of por olio value
that would be expected to be incurred a certain percentage of the me over a certain period of me
given assumed market condi ons.

• VaR requires the decomposi on of por olio performance into risk factors.

• The three methods of es ma ng VaR are the parametric method, the historical simula on method,
and the Monte Carlo simula on method.

• The parametric method of VaR es ma on typically provides a VaR es mate from the le tail of a
normal distribu on, incorpora ng the expected returns, variances, and covariances of the
components of the por olio.

• The parametric method exploits the simplicity of the normal distribu on but provides a poor
es mate of VaR when returns are not normally distributed, as might occur when a por olio contains
op ons.

• The historical simula on method of VaR es ma on uses historical return data on the por olio’s
current holdings and alloca on.

• The historical simula on method has the advantage of incorpora ng events that actually occurred
and does not require the speci ca on of a distribu on or the es ma on of parameters, but it is only
useful to the extent that the future resembles the past.

• The Monte Carlo simula on method of VaR es ma on requires the speci ca on of a sta s cal
distribu on of returns and the genera on of random outcomes from that distribu on.

• The Monte Carlo simula on method is extremely exible but can be complex and me consuming to
use.

• There is no single right way to es mate VaR.

• The advantages of VaR include the following: It is a simple concept; it is rela vely easy to understand
and easily communicated, capturing much informa on in a single number. It can be useful in
comparing risks across asset classes, por olios, and trading units and, as such, facilitates capital
alloca on decisions. It can be used for performance evalua on and can be veri ed by using
backtes ng. It is widely accepted by regulators.

• The primary limita ons of VaR are that it is a subjec ve measure and highly sensi ve to numerous
discre onary choices made in the course of computa on. It can underes mate the frequency of
extreme events. It fails to account for the lack of liquidity and is sensi ve to correla on risk. It is
vulnerable to trending or vola lity regimes and is o en misunderstood as a worst-case scenario. It
can oversimplify the picture of risk and focuses heavily on the le tail.
ti
ti
ti
ti
ti
ti
ti
ti
fl
ti
ti
tf
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
tf
tf
ti
ti
ti
fl
ti
ti
ti
ft
ti
ti
ti
ti
ft
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
fi
ti
ti
ti
ti
tf
ti
ft
tf
ti
tf
ti
ti
 • There are numerous varia ons and extensions of VaR, including condi onal VaR (CVaR), incremental
VaR (IVaR), and marginal VaR (MVaR), that can provide addi onal useful informa on.

• Condi onal VaR is the average loss condi onal on exceeding the VaR cuto .

• Incremental VaR measures the change in por olio VaR as a result of adding or dele ng a posi on
from the por olio or if a posi on size is changed rela- ve to the remaining posi ons.

• MVaR measures the change in por olio VaR given a small change in the por olio posi on. In a
diversi ed por olio, MVaRs can be summed to determine the contribu on of each asset to the
overall VaR.

• Ex ante tracking error measures the degree to which the performance of a given investment por olio
might deviate from its benchmark.

• Sensi vity measures quan fy how a security or por olio will react if a single risk factor changes.
Common sensi vity measures are beta for equi es; dura on and convexity for bonds; and delta,
gamma, and vega for op ons. Sensi vity measures do not indicate which por olio has greater loss
poten al.

• Risk managers can use deltas, gammas, vegas, dura ons, convexi es, and betas to get a
comprehensive picture of the sensi vity of the en re por olio.

• Stress tests apply extreme nega ve stress to a par cular por olio exposure.

• Scenario measures, including stress tests, are risk models that evaluate how a por olio will perform
under certain high-stress market condi ons.

• Scenario measures can be based on actual historical scenarios or on hypothe cal scenarios.

• Historical scenarios are scenarios that measure the por olio return that would result from a repeat of
a par cular period of nancial market history.

• Hypothe cal scenarios model the impact of extreme movements and co-movements in di erent
markets that have not previously occurred.

• Reverse stress tes ng is the process of stressing the por olio’s most signi cant exposures.

• Sensi vity and scenario risk measures can complement VaR. They do not need to rely on history, and
scenarios can be designed to overcome an assump on of normal distribu ons.

• Limita ons of scenario measures include the following: Historical scenarios are unlikely to re-occur in
exactly the same way. Hypothe cal scenarios may incorrectly specify how assets will co-move and
thus may get the magnitude of movements wrong. And, it is di cult to establish appropriate limits on
a scenario analysis or stress test.

• Constraints are widely used in risk management in the form of risk budgets, posi on limits, scenario
limits, stop-loss limits, and capital alloca on.

• Risk budge ng is the alloca on of the total risk appe te across sub-por olios.
ti
ti
ti
ti
ti
ti
fi
ti
ti
tf
tf
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
tf
ti
ti
ti
ti
ti
tf
ti
ti
ti
ti
ti
tf
ti
ti
tf
tf
ti
tf
ti
tf
ffi
ti
ti
ti
tf
fi
ti
ff
tf
ti
tf
ti
ti
ti
tf
ti
ti
ff
ti
tf
 • A scenario limit is a limit on the es mated loss for a given scenario, which, if exceeded, would require
correc ve ac on in the por olio.

• A stop-loss limit either requires a reduc on in the size of a por olio or its complete liquida on (when
a loss of a par cular size occurs in a speci ed period).

• Posi on limits are limits on the market value of any given investment.

• Risk measurements and constraints in and of themselves are not restric ve or unrestric ve; it is the
limits placed on the measures that drive ac on.

• The degree of leverage, the mix of risk factors to which the business is exposed, and accoun ng or
regulatory requirements in uence the types of risk measures used by di erent market par cipants.

• Banks use risk tools to assess the extent of any liquidity and asset/liability mismatch, the probability
of losses in their investment por olios, their over- all leverage ra o, interest rate sensi vi es, and the
risk to economic capital.

• Asset managers’ use of risk tools focuses primarily on vola lity, probability of loss, or the probability
of underperforming a benchmark.

• Pension funds use risk measures to evaluate asset/liability mismatch and surplus at risk.

• Property and casualty insurers use sensi vity and exposure measures to ensure exposures remain
within de ned asset alloca on ranges. They use economic capital and VaR measures to es mate the
impairment in the event of a catastrophic loss. They use scenario analysis to stress the market risks
and insurance risks simultaneously.

• Life insurers use risk measures to assess the exposures of the investment por olio and the annuity
liability, the extent of any asset/liability mismatch, and the poten al stress losses based on the
di erences between the assets in which they have invested and the liabili es resul ng from the
insurance contracts they have wri en.
ff
ti
ti
fi
ti
ti
ti
fl
tf
tf
tt
ti
ti
ti
fi
ti
ti
tf
ti
ti
ti
ff
ti
tf
ti
ti
ti
ti
ti
ti
ti
ti
 Learning Module 4 Backtes ng and Simula on

In this reading, we have discussed how to perform rolling-window backtes ng—a widely used technique
in the investment industry. We also described how to use scenario analysis and simula on along with
sensi vity analysis to supplement back- tes ng, so investors can be er account for the randomness in
data that may not be fully captured by backtes ng.

• The main objec ve of backtes ng is to understand the risk–return trade-o of an investment strategy
by approxima ng the real-life investment process.

• The basic steps in rolling-window backtes ng are specifying the investment hypothesis and goals,
determining the rules and processes behind an investment strategy, forming an investment por olio
according to those rules, rebalancing the por olio periodically, and compu ng the performance and
risk pro les of the strategy.

• In the rolling-window backtes ng methodology, researchers use a rolling-window (or walk-forward)

framework, t/calibrate factors or trade signals based on the rolling window, rebalance the por olio
periodically, and then track the performance over me. Thus, rolling-window backtes ng is a proxy
for actual inves ng.

• Analysts need to pay a en on to several behavioral issues in backtes ng, including survivorship bias
and look-ahead bias.

• Asset (and factor) returns are o en nega vely skewed and exhibit excess kurtosis (fat tails) and tail
dependence compared with a normal distribu on. As a result, standard rolling-window backtes ng
may be unable to fully account for the randomness in asset returns, par cularly on downside risk.

• Financial data o en face structural breaks. Scenario analysis can help investors understand the
performance of an investment strategy in di erent structural regimes.

• Historical simula on is rela vely straigh orward to perform but shares pros and cons similar to those
of rolling-window backtes ng. For example, a key assump on these methods share is that the
distribu on pa ern from the historical data is su cient to represent the uncertainty in the future.
Bootstrapping (or random draws with replacement) is o en used in historical simula on.

• Monte Carlo simula on is a more sophis cated technique than historical simula on. In Monte Carlo
simula on, the most important decision is the choice of func onal form of the sta s cal distribu on
of decision variables/ return drivers. Mul variate normal distribu on is o en used in investment
research, owing to its simplicity. However, a mul variate normal distribu on cannot account for
nega ve skewness and fat tails observed in factor and asset returns.

• Sensi vity analysis, a technique for exploring how a target variable and risk pro les are a ected by
changes in input variables, can further help investors understand the limita ons of conven onal
Monte Carlo simula on (which typically assumes a mul variate normal distribu on as a star ng
point).

• A mul variate skewed t-distribu on considers skewness and kurtosis but requires es ma on of more
parameters and thus is more likely to su er from larger es ma on errors.
ti
ti
ti
ti
ti
fi
ti
fi
ti
tt
ti
ti
ft
ti
ti
ti
tt
ti
ti
ti
ti
ti
ti
ft
ti
tf
ff
ti
ti
ti
ti
ti
ti
ff
tf
ti
ti
ti
ffi
ti
ti
ft
ti
ti
ti
ti
ti
tt
ti
ti
ti
ft
ff
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
tf
tf
ti
ti
 Learning Module 5 Economics and Investment Markets

In this reading, we have sought to explain the fundamental connec on between the prices of nancial
assets and the underlying economy. The connec on should be strong because ul mately all nancial
assets represent a claim on the real economy. Because all nancial assets o er a means of deferring
consump on, to make the connec on tangible we have explored the rela onship between these asset
prices and the consump on and saving decisions of economic agents.

• At any point in me, the market value of any nancial security is simply the sum of discounted values
of the cash ows that the security is expected to produce. The ming and magnitude of these
expected cash ows will thus be an integral part of the security’s market value, as will the discount
rate applied to these expected cash ows, which is the sum of a real default-free interest rate,
expected in a on, and possibly several risk premiums. Each of these elements will be in uenced by
the business cycle. It is through these components that the real economy exerts its in uence on the
market value of nancial instruments.

• The average level of real short-term interest rates is posi vely related to the trend rate of growth of
the underlying economy and also to the vola lity of economic growth in the economy. Other things
being equal, these rela onships mean that we should expect to nd that the average level of real
short-term interest rates is higher in an economy with high and vola le growth and lower in an
economy with lower, more stable growth.

• On average, over me, according to the Taylor rule, a central bank’s policy rate should comprise the
sum of an economy’s trend growth plus in a on expecta ons, which might, in turn, be anchored to
an explicit in a on target. This policy rate level is referred to as the neutral rate. Other things being
equal, when in a on is above (below) the targeted level, the policy rate should be above (below) the
neutral rate, and when the output gap is posi ve (nega ve), the policy rate should also be above
(below) the neutral rate. The policy rate can thus vary over me with in a on expecta ons and the
economy’s output gap.

• Short-term nominal rates will be closely related to a central bank’s policy rate of interest and will
comprise the real interest rate that is required to balance the requirements of savers and investors
plus investors’ expecta ons of in a on over the relevant borrowing or lending period. Short-term
nominal interest rates will be posi vely related to short-term real interest rates and to in a on
expecta ons.

• If bond investors were risk neutral, then the term structure of interest rates would be determined by
short-term interest rate expecta ons. But bond investors are risk averse, which means that they will
normally demand a risk premium for inves ng in even default-free government bonds. This risk
premium will generally rise with the maturity of these bonds because longer-dated government
bonds tend to be less nega vely correlated with consump on and, therefore, represent a less useful
consump on hedge for investors. Overall, the shape of the curve will be determined by a
combina on of short-term interest rates and in a on expecta ons as well as risk premiums. In turn,
these factors will be in uenced by the business cycle and policymakers.

• The yield di eren al between default-free conven onal government bonds and index-linked
equivalents will be driven by in a on expecta ons and a risk premium. The risk premium will be
largely in uenced by investors’ uncertainty about future in a on.
ti
ti
fl
ti
ti
ff
fl
fl
fl
ti
fl
fl
ti
ti
fi
ti
ti
ti
fl
ti
ti
ti
ti
fl
ti
fl
ti
ti
ti
ti
fl
fl
ti
ti
ti
fi
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
fi
fl
ti
ti
ti
ti
fi
ti
ti
fl
ti
ti
ff
ti
fl
ti
fl
fl
ti
fi
fi
 • The di erence between the yield on a corporate bond and that on a government bond with the same
currency denomina on and maturity is referred to as the measured credit spread. It is conceptually
akin (but not equal) to the risk premium demanded by investors in compensa on for the addi onal
credit risk that they bear compared with that embodied in the default-free government bond. It tends
to rise in mes of economic weakness, as the probability of default rises, and tends to narrow in
mes of robust economic growth, when defaults are less common.

• The uncertainty about and me varia on in future equity cash ows (dividends) is a dis nct feature
of equity investment, as opposed to corporate bond investment. This feature explains why we would
expect the equity premium to be larger than the credit premium. In mes of economic weakness or
stress, the uncertainty about future dividends will tend to be higher, and we should thus expect the
equity risk premium to rise in such an economic environment.

• Given the uncertain nature of the cash ows generated by equi es, investors will demand an equity
risk premium because the consump on hedging proper es of equi es are poor. In other words,
equi es tend not to pay o in bad mes. Because in the event of company failure an equity holder
will lose all of his or her investment whereas an investor in the company’s bonds may recover a
signi cant por on of his or her investment, it would be reasonable to assume that a risk-averse
investor would demand a higher premium on an equity holding than on a corporate bond holding.
The two premiums will tend to be posi vely correlated over me and will tend to be in uenced by
the business cycle in similar ways.

• The P/E tends to rise during periods of economic expansion and to fall during recessions. A “high” P/E
could be the result of a number of factors, including the following: falling real interest rates, a decline
in the equity risk premium, an increase in the expecta on of future real earnings growth, an
expecta on of lower opera ng and/or nancial risk, or a combina on of all of these factors. All of
these components will be in uenced by the business cycle.

• The market value of an investment in commercial property can be derived in much the same way as
the market value of an investment in equity. The cash ows come in the form of rent, which can be
enhanced with addi onal redevelopment values as leases on proper es expire. These cash ows are
uncertain, and the uncertainty surrounding them will tend to rise when the economy turns down. We
might thus expect the risk premium demanded on commercial property investments to rise in these
mes.

• The pro-cyclical nature of commercial property prices means that investors will generally demand a
rela vely high risk premium in return for invest- ing in this asset class. The reason is that commercial
property is not a very good hedge against bad economic outcomes. In addi on, the illiquid nature of
property investment means that investors may also demand a liquidity premium for inves ng in this
asset class.
ti
ti
ti
ti
fi
ff
ti
ti
ti
ti
ti
ff
ti
ti
fl
ti
ti
ti
ti
fi
fl
ti
fl
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
fl
ti
ti
fl
ti
 Learning Module 6 Analysis of Ac ve Por olio Management

We have covered a number of key concepts and principles associated with ac ve port- folio
management. Ac ve management is based on the mathema cs and principles of risk and return from
basic mean–variance por olio theory but with a focus on value added compared with a benchmark
por olio. Cri cal concepts include the following:

• Value added is de ned as the di erence between the return on the managed por olio and the return
on a passive benchmark por olio. This di erence in returns might be posi ve or nega ve a er the
fact but would be expected to be posi ve before the fact or ac ve management would not be
jus ed.

• Value added is related to ac ve weights in the por olio, de ned as di erences between the various
asset weights in the managed por olio and their weights in the benchmark por olio. Individual
assets can be overweighted (have posi ve ac ve weights) or underweighted (have nega ve ac ve
weights), but the complete set of ac ve weights sums to zero.

• Posi ve value added is generated when posi ve-ac ve-weight assets have larger returns than
nega ve-ac ve-weight assets. By de ning individual asset ac ve returns as the di erence between
the asset total return and the benchmark return, value added is shown to be posi ve if and only if
end-of-period realized ac ve asset returns are posi vely correlated with the ac ve asset weights
established at the beginning of the period.

• Value added can come from a variety of ac ve por olio management deci- sions, including security
selec on, asset class alloca on, and even further decomposi ons into economic sector weigh ngs
and geographic or country weights.

• The Sharpe ra o measures reward per unit of risk in absolute returns, whereas the informa on ra o
measures reward per unit of risk in bench- mark rela ve returns. Either ra o can be applied ex ante
to expected returns or ex post to realized returns. The informa on ra o is a key criterion on which to
evaluate ac vely managed por olios.

• Higher informa on ra o por olios can be used to create higher Sharpe ra o por olios. The op mal
amount of ac ve management that maximizes a por olio’s Sharpe ra o is posi vely related to the
assumed forecas ng accuracy or ex ante informa on coe cient of the ac ve strategy.

• The ac ve risk of an ac vely managed strategy can be adjusted to its desired level by combining it
with a posi on in the benchmark. Furthermore, once an investor has iden ed the maximum Sharpe
ra o por olio, the total vola lity of a por olio can be adjusted to its desired level by combining it
with cash (two-fund separa on concept).

• The fundamental law of ac ve por olio management began as a conceptual framework for
evalua ng the poten al value added of various investment strategies, but it has also emerged as an
opera onal system for measuring the essen al components of those ac ve strategies.

• Although the fundamental law provides a framework for analyzing investment strategies, the
essen al inputs of forecasted asset returns and risks s ll require judgment in formula ng the
expected returns.
ti
ti
tf
ti
fi
ti
ti
ti
ti
ti
ti
tf
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
tf
ti
ti
ti
ti
ti
tf
ti
tf
tf
ff
tf
ti
tf
ti
fi
ti
ti
ff
tf
tf
ti
ti
ti
ti
ti
tf
tf
ti
ti
tf
ti
ti
ffi
fi
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
tf
tf
tf
ff
ti
ti
ti
ti
ft
ti
ti
ti
ti
ti
 • The fundamental law separates the expected value added, or por olio return rela ve to the
benchmark return, into the basic elements of the strategy:
skill as measured by the informa on coe cient,
structuring of the por olio as measured by the transfer coe cient,
breadth of the strategy measured by the number of independent deci- sions per year, and
aggressiveness measured by the benchmark tracking risk.

• The last three of these four elements may be beyond the control of the investor if they are speci ed
by investment policy or constrained by regula on.

• The fundamental law has been applied in se ngs that include the selec on of country equity
markets in a global equity fund and the ming of credit and dura on exposures in a xed-income
fund.

• The fundamental law of ac ve management has limita ons, including uncertainty about the ex ante
informa on coe cient and the conceptual de ni on of breadth as the number of independent
decisions by the investor.
ti
ffi
tf
ti
ti
ti
ffi
tti
ti
fi
ti
ti
ffi
ti
tf
ti
ti
fi
fi
 Learning Module 7 Trading Costs and Electronic Markets

This reading explains the implicit and explicit costs of trading as well as widely used methods for
es ma ng transac on costs. The reading also describes developments in electronic trading, the main
types of electronic traders, their needs for speed and ways in which they trade. Electronic trading
bene ts investors through lower transac on costs and greater e ciencies but also introduces systemic
risks and the need to closely monitor markets for abusive trading prac ces. Appropriate market
governance and regulatory policies will help reduce the likelihood of events such as the 2010 Flash
Crash. The reading’s main points include:

• Dealers provide liquidity to buyers and sellers when they take the other side of a trade if no other
willing traders are present.

• The bid–ask spread is the di erence between the bid and the ask prices. The e ec ve spread is two
mes the di erence between the trade price and the midquote price before the trade occurred. The
e ec ve spread is a poor es mate of actual transac on costs when large orders have been lled in
many parts over me or when small orders receive price improvement.

• Transac on costs include explicit costs and implicit costs. Explicit costs are the direct costs of trading.
They include broker commissions, transac on taxes, stamp du es, and exchange fees. Implicit costs
include indirect costs, such as the impact of the trade on the price received. The bid–ask spread,
market impact, delay, and un lled trades all contribute to implicit trading costs.

• The implementa on shor all method measures the total cost of implement- ing an investment
decision by capturing all explicit and implicit trading costs. It includes the market impact costs, delay
costs, as well as opportunity costs.

• The VWAP method of es ma ng transac on costs compares average ll prices to average market
prices during a period surrounding the trade. It tends to produce lower transac on cost es mates
than does implementa on shor all because it o en does not measure the market impact of an order
well.

• Markets have become increasingly fragmented as venues trading the same instruments have
proliferated. Trading in any given instrument now occurs in mul ple venues.

• The advantages of electronic trading systems include cost and opera onal e ciencies, lack of human
bias, extraordinarily fast speed, and in nite span and scope of a en on.

• Latency is the elapsed me between the occurrence of an event and a sub- sequent ac on that
depends on that event. Traders use fast communica on systems and fast computer systems to
minimize latency to execute their strategies faster than others.

• Hidden orders, quote leapfrogging, ickering quotes, and the use of machine learning to support
trading strategies commonly are found in electronic markets.

• Traders commonly use advanced order types, trading tac cs, and algorithms in electronic markets.

• Electronic trading has bene ted investors through greater trade process e ciencies and reduced
transac on costs. At the same me, electronic trading has increased systemic risks.
ti
ff
ti
ti
fi
ti
ti
ti
ff
ti
ti
ti
ti
ti
ti
tf
fi
ff
ti
ti
fi
ti
tf
fl
fi
ti
ti
ti
ft
ti
ti
ti
ti
ti
tt
ffi
ti
ti
fi
ti
ffi
ffi
ff
ti
ti
ti
ti
fi
 • Examples of systemic risks posed by electronic traders include: runaway algorithms that produce
streams of unintended orders caused by program- ming mistakes, fat nger errors that occur when a
manual trader submits a larger order than intended, overlarge orders that demand more liquidity
than the market can provide, and malevolent order streams created deliberately to disrupt the
markets.

• Real- me surveillance of markets o en can detect order front running and various market
manipula on strategies.

• Market manipulators use such improper ac vi es as trading for market impact, rumormongering,
wash trading, and spoo ng to further their schemes.

• Market manipula on strategies include blu ng, squeezing, cornering, and gunning.
ti
ti
ti
fi
ft
ti
ffi
ti
fi
 Ethical and Professional Standards
Learning Module 1 Code of Ethics and Standards of Professional Conduct

The Code of Ethics

• Act with integrity, competence, diligence, and respect and in an ethical manner with the public,
clients, prospec ve clients, employers, employees, colleagues in the investment profession, and other
par cipants in the global capital markets.

• Place the integrity of the investment profession and the interests of clients above their own personal
interests.
• Use reasonable care and exercise independent professional judgment when conduc ng investment
analysis, making investment recommenda ons, taking investment ac ons, and engaging in other
professional ac vi es.

• Prac ce and encourage others to prac ce in a professional and ethical manner that will re ect credit
on themselves and the profession.

• Promote the integrity and viability of the global capital markets for the ul mate bene t of society.

• Maintain and improve their professional competence and strive to maintain and improve the
competence of other investment professionals.

Standards of Professional Conduct

I. PROFESSIONALISM

A. Knowledge of the Law

• Members and Candidates must understand and comply with all applicable laws, rules, and regula ons
(including the CFA Ins tute Code of Ethics and Standards of Professional Conduct) of any government,
regulatory organiza on, licensing agency, or professional associa on governing their professional
ac vi es. In the event of con ict, Members and Candidates must comply with the more strict law,
rule, or regula on. Members and Candidates must not knowingly par cipate or assist in and must
dissociate from any viola on of such laws, rules, or regula ons.

B. Independence and Objec vity

• Members and Candidates must use reasonable care and judgment to achieve and maintain
independence and objec vity in their professional ac vi es. Members and Candidates must not o er,
solicit, or accept any gi , bene t, compensa on, or considera on that reasonably could be expected
to compromise their own or another’s independence and objec vity.
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ft
ti
ti
ti
fl
fi
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
fl
ti
ff
 C. Misrepresenta on

• Members and Candidates must not knowingly make any misrepresenta ons rela ng to investment
analysis, recommenda ons, ac ons, or other professional ac vi es.

D. Misconduct

• Members and Candidates must not engage in any professional conduct involving dishonesty, fraud, or
deceit or commit any act that re ects adversely on their professional reputa on, integrity, or
competence.

II. INTEGRITY OF CAPITAL MARKETS

A. Material Nonpublic Informa on

• Members and Candidates who possess material nonpublic informa on that could a ect the value of
an investment must not act or cause others to act on the informa on.

B. Market Manipula on

• Members and Candidates must not engage in prac ces that distort prices or ar cially in ate trading
volume with the intent to mislead market par cipants.

III. DUTIES TO CLIENTS

A. Loyalty, Prudence, and Care

• Members and Candidates have a duty of loyalty to their clients and must act with reasonable care
and exercise prudent judgment. Members and Candidates must act for the bene t of their clients and
place their cli- ents’ interests before their employer’s or their own interests.

B. Fair Dealing

• Members and Candidates must deal fairly and objec vely with all clients when providing investment
analysis, making investment recommenda ons, taking investment ac on, or engaging in other
professional ac vi es.

C. Suitability

• When Members and Candidates are in an advisory rela onship with a client, they must:
Make a reasonable inquiry into a client’s or prospec ve client’s investment experience, risk and
return objec ves, and nancial constraints prior to making any investment recommenda on or
taking investment ac on and must reassess and update this informa on regularly.
Determine that an investment is suitable to the client’s nancial situa on and consistent with the
client’s wri en objec ves, man- dates, and constraints before making an investment
recommenda on or taking investment ac on.
Judge the suitability of investments in the context of the client’s total por olio.
tt
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
ti
ti
ti
ti
ti
tf
ti
ti
fi
fi
ti
ff
ti
fl
 • When Members and Candidates are responsible for managing a por olio to a speci c mandate,
strategy, or style, they must make only investment recommenda ons or take only investment ac ons
that are consistent with the stated objec ves and constraints of the por olio.

D. Performance Presenta on

• When communica ng investment performance informa on, Members and Candidates must make
reasonable e orts to ensure that it is fair, accurate, and complete.

E. Preserva on of Con den ality

• Members and Candidates must keep informa on about current, former, and prospec ve clients
con den al unless:
The informa on concerns illegal ac vi es on the part of the client or prospec ve client,
Disclosure is required by law, or
The client or prospec ve client permits disclosure of the informa on.

IV. DUTIES TO EMPLOYERS

A. Loyalty

• In ma ers related to their employment, Members and Candidates must act for the bene t of their
employer and not deprive their employer of the advantage of their skills and abili es, divulge
con den al informa on, or otherwise cause harm to their employer.

B. Addi onal Compensa on Arrangements

• Members and Candidates must not accept gi s, bene ts, compensa on, or considera on that
competes with or might reasonably be expected to create a con ict of interest with their employer’s
interest unless they obtain wri en consent from all par es involved.

C. Responsibili es of Supervisors

• Members and Candidates must make reasonable e orts to ensure that anyone subject to their
supervision or authority complies with applicable laws, rules, regula ons, and the Code and
Standards.

V. INVESTMENT ANALYSIS, RECOMMENDATIONS, AND ACTIONS

A. Diligence and Reasonable Basis

Members and Candidates must:

• Exercise diligence, independence, and thoroughness in analyzing investments, making investment
recommenda ons, and taking investment ac ons.
• Have a reasonable and adequate basis, supported by appropriate research and inves ga on, for any
investment analysis, recommenda on, or ac on.
fi
fi
ti
tt
ti
ti
ti
ti
ti
ff
ti
ti
fi
ti
ti
ti
ti
ti
tt
ti
ti
ti
ti
ti
ti
ft
ti
ff
fi
ti
ti
fl
ti
ti
tf
ti
ti
tf
ti
ti
fi
ti
ti
ti
ti
fi
ti
 B. Communica on with Clients and Prospec ve Clients

Members and Candidates must:

• Disclose to clients and prospec ve clients the basic format and general principles of the investment
processes they use to analyze investments, select securi es, and construct por olios and must
promptly disclose any changes that might materially a ect those processes.
• Disclose to clients and prospec ve clients signi cant limita ons and risks associated with the
investment process.
• Use reasonable judgment in iden fying which factors are important to their investment analyses,
recommenda ons, or ac ons and include those factors in communica ons with clients and
prospec ve clients.
• Dis nguish between fact and opinion in the presenta on of investment analysis and
recommenda ons.

C. Record Reten on

• Members and Candidates must develop and maintain appropriate records to support their
investment analyses, recommenda ons, ac ons, and other investment-related communica ons with
clients and prospec ve clients.

VI. CONFLICTS OF INTEREST

A. Disclosure of Con icts

• Members and Candidates must make full and fair disclosure of all ma ers that could reasonably be
expected to impair their independence and objec vity or interfere with respec ve du es to their
clients, prospec ve clients, and employer. Members and Candidates must ensure that such
disclosures are prominent, are delivered in plain language, and communicate the relevant
informa on e ec vely.

B. Priority of Transac ons

• Investment transac ons for clients and employers must have priority over investment transac ons in
which a Member or Candidate is the bene cial owner.

C. Referral Fees

• Members and Candidates must disclose to their employer, clients, and prospec ve clients, as
appropriate, any compensa on, considera on, or bene t received from or paid to others for the
recommenda on of products or services.

VII.RESPONSIBILITIES AS A CFA INSTITUTE MEMBER OR CFA CANDIDATE

A. Conduct as Par cipants in CFA Ins tute Programs

• Members and Candidates must not engage in any conduct that compromises the reputa on or
integrity of CFA Ins tute or the CFA designa on or the integrity, validity, or security of CFA Ins tute
programs.
ti
ti
ti
ti
ti
ti
ti
ff
ti
ti
ti
ti
fl
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
ti
fi
ti
ti
ti
fi
ti
ti
ff
fi
ti
ti
tt
ti
tf
ti
ti
ti
ti
ti
ti
ti
 B. Reference to CFA Ins tute, the CFA Designa on, and the CFA Program

• When referring to CFA Ins tute, CFA Ins tute membership, the CFA designa on, or candidacy in the
CFA Program, Members and Candidates must not misrepresent or exaggerate the meaning or
implica ons of membership in CFA Ins tute, holding the CFA designa on, or candidacy in the CFA
Program.
ti
ti
ti
ti
ti
ti
ti
ti