Practice Final Exam I

FM423
Asset Markets

Part B
Question 2. (25 marks)

(a) Consider the two (excess return) one-factor-model regression results for stock A and B.
The risk-free rate over the period was 6%, and the market’s average return was 14%.
Performance is measured using a market-factor model regression on excess returns.

Stock A Stock B
One-factor model regression estimates 1%+1.2(rm − rf ) 2%+0.8(rm − rf )
Residual standard deviation, σ(e) 10.5% 19.4%
Standard deviation of excess return 21.4% 25.5%

(i) (4 marks) Calculate the following statistics for each stock: alpha, appraisal ratio,
Sharpe ratio, and Treynor ratio.
Solution: A’s alpha is 0.01 and B’s alpha is 0.02.
Appraisal ratio for A: 0.01/0.105 = 0.095.
Appraisal ratio for B: 0.02/0.194 = 0.103.
Sharpe ratio for A: [0.01 + 1.2 (0.14 − 0.06)] /0.214 = 0.106/0.214 = 0.495.
Sharpe ratio for B: [0.02 + 0.8 (0.14 − 0.06)] /0.255 = 0.084/0.255 = 0.329.
Treynor ratio for A: 0.106/1.2 = 0.088.
Treynor ratio for B: 0.084/0.8 = 0.105.
(ii) Which stock is the best choice under the following circumstances and why?
i. (2 marks) This is the only risky asset to be held by the investor.
Solution: From this strategy perspective, return to total risk dictates use of
the Sharpe ratio. Therefore A, since it has the highest Sharpe ratio.
ii. (2 marks) This stock will be mixed with the rest of the investor’s portfolio,
currently composed solely of holdings in the market index fund. This stock’s
weight in the resulting portfolio is large in relation to the weights of the stocks
in the market portfolio.
Solution: If a single stock is to be combined with an efficiently diversified
portfolio (the market portfolio), and the stock’s weight in the resulting over-
all portfolio is large in relation to the weights of the stocks in the market

c LSE FM423 Page 1 of 10
 portfolio, the stock’s unsystematic risk becomes relevant, and the selection’s
contribution should be based on the Appraisal Ratio. Therefore B, since it
has the highest appraisal ratio.
iii. (2 marks) This is one of many stocks that the investor is analyzing to form
an actively managed stock portfolio.
Solution: If a single stock is to be added into an efficiently diversified, non-
market index portfolio and the stock’s weight is comparable to the weights
of the other assets, its unsystematic risk becomes irrelevant and the Treynor
ratio is appropriate. Therefore B, since it has the highest Treynor ratio.
(iii) (3 marks) You are asked to construct a portfolio that invests exclusively in stocks
A and B. Moreover, the return of this portfolio should be uncorrelated with the
market return (that is, the beta of the portfolio is zero). Give the weights of
such a portfolio and compute its alpha and its Sharpe ratio assuming that the
idiosyncratic risks of A and B are independent.
Solution: If the portfolio is uncorrelated with the market then its beta is zero:

βp = (wA ) (1.2) + (1 − wA ) (0.8) = 0

wA = −0.8/ (1.2 − 0.8) = −2.

Therfore the weight in A is -2 and the weight in B is 3. The alpha of the portfolio
is
αp = (−2) (0.01) + (3) (0.02) = 0.04.
The Sharpe ratio of the portfolio is determined by

E (p) − rf
Sp = q .
2
β 2 σM + σe2p

Since the beta of the portfolio is zero the Sharpe ratio is

0.1 − 0.06
Sp = q = 0.065.
(−2)2 (0.105)2 + (3)2 (0.194)2

(b) Give brief answers to the following questions:

(i) (2 marks) What are the main assumptions and the logic of APT?
Solution: Absence of arbitrage; APT is the pricing equation for well-diversified
portfolios (no idiosyncratic risks); (using existing assets to create) factor replicat-
ing portfolios, and by no arbitrage principle to pin down factor premiums; the
expected return for the asset is related to its factor-loadings, factor premiums.

c LSE FM423 Page 2 of 10
 (ii) (2 marks) What are the advantages and disadvantages of APT relative to the
CAPM?
Solution: More factors, capture more sources of systematic risks; less restrictive
assumptions; but offers no guidance on what these factors are.
(iii) (2 marks) Does behavioral finance dispute any of the following statements? Why?
i. No investment strategy can earn excess risk-adjusted returns greater than
what are warranted for its risk level.
Solution: It does not dispute the statement “No investment strategy can
earn excess risk-adjusted returns greater than what are warranted for its risk
level.” – due to limits to arbitrage.
ii. Market prices are equal to the fundamental value of the asset (prices are set
by rational agents).
Solution: It disputes the statement “Market prices is equal to the fundamental
value of the asset (prices are set by rational agents).”
(iv) (2 marks) As discussed in the lectures, noise trader risk is an important deterrent
to arbitrage activity. What are the main assumptions behind noise trader risk?
Solution: Though mispricing should eventually disappear in the long run, many
real-world arbitrageurs likely have a short horizon since they manage other peo-
ple’s money. Reasons for the short horizon include impatient and uninformed
investors in funds dictating the investment horizon and creditors calling back
loans as the loss against the collateral grows in the short term.
(c) (4 marks) Consider a forward contract to buy GBP with USD in T years. Let S denote
the GBP/USD spot exchange rate (i.e., $1 = £S); F denote the forward price of GBP
(i.e., $1 = £F ); rU K and rU S denote the annualized T-year risk-free rates in the UK
and the US. Derive F as a function of S, rU K and rU S .
Solution: Consider two investment investment strategies by a US investor. In each
strategy the investor starts with $1:
Strategy 1: At t = 0 convert the $1 into GBP using the current spot rate SGBP/U SD
and invest for T years at the UK risk-free rate. At time T the investor would have in
GBP, SGBP/U SD (1 + rU K )T .

Strategy 2: At t = 0 invest the $1 for T years at the US risk-free rate and simulta-
neously at t = 0 enter a forward contract to buy GBP paying $1 (1 + rU S )T . At time
T the investor would have in GBP, FGBP/U SD (1 + rU S )T .
Since both investors started off with $1 and all rates are known at t = 0 and assumed
to be risk-free then by no arbitrage both investments will have the same payoff in GBP
at t = T :
SGBP/U SD (1 + rU K )T = FGBP/U SD (1 + rU S )T

(1 + rU K )T
FGBP/U SD = SGBP/U SD .
(1 + rU S )T

c LSE FM423 Page 3 of 10
Question 3. (25 marks)
(a) Consider a two-period binomial model (t = {0, 1, 2}), with a risky non-dividend paying
stock and a risk-free bank account. Assume that the price of the stock is now S0 =
£100 and in the first period it can go up by 20% or down by 10% (so u = 1.2 and
d = 0.9). The physical probability that the stock moves up is 0.6. The same behaviour
is exhibited also in the second period. The risk-free rate is r1 = 2% in the first period,
and r2 = 5% in the second period (so that £1 deposited in the bank at t = 0 will be
£1.02 at t = 1, and £1.02 × 1.05 = £1.071 at t = 2).

(i) (4 marks) What is the price at t = 0 of a European call option on the stock with
maturity at t = 2 and strike price K = 110. Does the price at t = 0 change if the
call option is American? Explain.
Solution: For the risk neutral probabilities we have q1 = (1.02 − 0.9)/(1.2 −
0.9) = 0.4, and q2 = (1.05 − 0.9)/(1.2 − 0.9) = 0.5. For the stock prices we have
Su = 120, Sd = 90, Suu = 144, Sud = 108, Sdu = 108 and Sdd = 81. The payoffs
at maturity for the European Call are Cuu = 34, Cud = 0, Cdu = 0, and Cdd = 0.
Hence,
q2 Cuu + (1 − q2 )Cud
Cu = = (0.5 ∗ 34)/1.05 = 16.190,
1 + r2
q2 Cud + (1 − q2 )Cdd
Cd = = 0,
1 + r2
q1 Cu + (1 − q1 )Cd
C = = (0.4 ∗ 16.190)/1.02 = 6.349.
1 + r1
Early exercise is never optimal for an American Call option on a non-dividend
paying stock. Hence the European and American Call options have the same price
of £6.349.
(ii) (4 marks) What is the price at t = 0 of a European put option on the stock with
maturity at t = 2 and strike price K = 110. Does the price at t = 0 change if the
put option is American? Explain.
Solution: The payoffs at maturity for the European Put are Puu = 0, Pud =
2, Pdu = 2, and Pdd = 29. Hence,
q2 Puu + (1 − q2 )Pud
Pu = = 0.5 ∗ 2/1.05 = 0.952,
1 + r2
q2 Pud + (1 − q2 )Pdd
Pd = = (0.5 ∗ 2 + 0.5 ∗ 29)/1.05 = 14.762,
1 + r2
q1 Pu + (1 − q1 )Pd
P = = (0.4 ∗ 0.952 + 0.6 ∗ 14.762)/1.02 = 9.057.
1 + r1
The price at t = 0 of a European Put option is £9.057. The American Put option
would be exercised early in the down node (20 > 14.762) and its value at t = 0
would be (0.4*0.952 + 0.6*20)/1.02 = £12.138.

c LSE FM423 Page 4 of 10
 (iii) (4 marks) What is the price at t = 0 of an Average Strike Asian call option on
the stock with maturity at t = 2. The payoff at maturity is [ST − Save ]+ where
Save is the average price the stock reaches during the life of the option. An Asian
call option can be only exercised at maturity.
Solution: For the average stock prices we have Suu = (100 + 120 + 144)/3 =
121.333, Sud = (100 + 120 + 108)/3 = 109.333, Sdu = (100 + 90 + 108)/3 = 99.333
and Sdd = (100 + 90 + 81)/3 = 90.333. The payoffs at maturity for the Average
Strike Asian Call are Cbuu = 22.667, Cbud = 0, C
bdu = 8.667, and C bdd = 0. Hence,

bu = q2 Cuu + (1 − q2 )Cud = (0.5 ∗ 22.667)/1.05 = 10.794,

b b
C
1 + r2
bd = q2 Cdu + (1 − q2 )Cdd = (0.5 ∗ 8.667)/1.05 = 4.127,
b b
C
1 + r2
b = q1 Cu + (1 − q1 )Cd = (0.4 ∗ 10.794 + 0.6 ∗ 4.127)/1.02 = 6.661.
b b
C
1 + r1
The price at t = 0 of a Average Strike Asian Call is £6.661.

(b) Now consider a stock with Geometric Brownian motion dynamics as in the Black &
Scholes model:

dSt
= µdt + σdWt ,
St
µ and σ are positive known constants. There is also a money-market account with
dynamics

dBt
= rdt,
Bt
where r is a positive known constant.

(i) (4 marks) What is the sign of ρ (i.e., the sensitivity of the price with respect to
r) and ν (i.e., the sensitivity of the price with respect to σ) for a European call
and for a European put in the Black & Scholes model? Explain.
Solution: For a European Call ρ is positive because higher r means a lower present
value of the perceived cost K. For a European put ρ is negative because higher r
means a lower present value of the perceived gain K. For both a European Call
and a European Put ν is positive because higher volatility σ may only lead to
higher payoffs, since the downside is always bounded at zero, this is the ‘cherry
picking’ feature of options.

c LSE FM423 Page 5 of 10
 (ii) (4 marks) Consider a stock where µ = 10% and σ = 20%. The current stock price
is £100. What is the probability that a European call option on the stock with
a strike price of £120 and an expiration date in 3 years will be exercised? Leave
your answer in term of the standard normal cumulative distribution function, Φ.
Solution:
ln(ST ) ∼ N (ln(S0 ) + (µ − σ 2 /2)T, σ 2 T )
ln(ST ) ∼ N (ln(100) + (0.1 − 0.22 /2)(3), (0.22 )(3))
ln(ST ) ∼ N (4.845, 0.12)
The European Call will be exercised if S3 ≥ 120. We can also write this as
ln(S3 ) ≥ ln(120). Thus we want

Pr [ln(S3 ) ≥ ln(120)]
h √ i
Pr Z ≥ (ln(120) − 4.845)/( 0.12)

Pr [Z ≥ −0.167]
which is 1 − Φ(−0.167) which can also be expressed as Φ(0.167).
(iii) (5 marks) Consider a derivative that at maturity date T has a payoff of ST10 , i.e.
the 10th power of the stock price at maturity. Derive the price of this derivative
at t = 0, using the risk-neutral valuation method.
Solution:
V0 = e−rT EQ0 [VT ]

V0 = e−rT E0 ST10
Q
 

Under the risk neutral probability Q, we know that

h i
ST = S0 exp r − σ 2 /2 T + σ W


fT

so that h i
ST10 = S010 exp 10r − 5σ 2 T + 10σ W


fT

Hence regarding the risk neutral expectation we have

h h ii
EQ ST10 = EQ S010 exp 10r − 5σ 2 T + 10σ W
 

fT

h i
EQ ST10 = S010 exp 10r − 5σ 2 T EQ exp[10σ W
  

fT ]
 
 10  10
 2

 1 2
E ST = S0 exp 10r − 5σ T exp (10σ) T
Q
2
EQ ST10 = S010 exp 10r − 5σ 2 T exp 50σ 2 T
  
  

c LSE FM423 Page 6 of 10
 EQ ST10 = S010 exp 10r + 45σ 2 T
  


The price of this derivative today is

V0 = exp (−rT ) S010 exp 10r + 45σ 2 T




V0 = S010 exp 9r + 45σ 2 T




c LSE FM423 Page 7 of 10
Question 4. (25 points)
In this question you are first asked to prove two no arbitrage restrictions. All options are
written on the same non-dividend paying stock, have the same strike price K, and have the
same maturity T . [Notes: Make sure every step in your proof is justified.]

(i) (2 points) CtEU = PtEU + St − PVTt (K) (i.e., Put-Call Parity),

Solution: CTEU = [ST − K]+ = PTEU + ST − K. Students can show this either by showing
the payoff diagram or solving for the payoff function of the replicating portfolio (as
we did in lecture 15) [4 points]. Then, given no arbitrage, the present value of the
left-hand side must equal that of the right hand side [1 point].

(ii) (3 points) CtAM > St −K (i.e., early exercise of an American call option is never optimal
on a non-dividend paying stock),
Solution: CtAM ≥ CtEU = PtEU + St − PVTt (K) > St − PVTt (K) > St − K [5 points].
So it’s never optimal to exercise early. The first inequality is based on the fact that
American options are worth at least as much as otherwise identical European options.
The equality follows put-call parity.

for all 0 ≤ t < T , where the subscript t denotes the time at which the corresponding
option/stock price is quoted; the superscripts AM, EU denote that the corresponding option
is American, European. PVTt (·) is the present value operator bringing money risk-free from
T to t.

(iii) (2 points) Discuss whether (ii) can be generalized for American call options on com-
modities. Explain your answer in terms of whether a commodity is in contango or
backwardation.
Solution: It can be generalized for commodities for which the convenience yield is lower
than the storage cost, i.e., for commodities in contango [5 points]. This can be seen
by the proof where at maturity we ‘pay’ the future value of storage cost (assume the
convenience yield is zero) and hence end up receiving an even higher positive payoff.

(iv) (3 points) Now, assume that the financial stock in (ii) pays dividends. The (risk-free)
value of dividends at T for someone who bought the stock at t < T is Dt > 0. Find a
sufficient condition between K and Dt so that the inequality in (ii) still holds.
Solution: Follow the same steps as in the proof of (ii). CtAM ≥ CtEU = PtEU + St −
Dt − PVTt (K) > St − Dt − PVTt (K). To ensure that this is greater than St − K, the
sufficient condition is K − PVTt (K) > Dt [5 points].

(v) (5 points) An asset-or-nothing digital call option with strike price K and maturity date
T has payoff ST if ST > K, and 0 otherwise, where ST is the price of the underlying at
maturity. Price this option for strike price 60 and time-to-maturity 2 years, assuming
that the price of the underlying evolves according to the binomial tree in Figure 1

c LSE FM423 Page 8 of 10
 below (one binomial step per year; the numbers along the branches are the physical
probabilities of an up/down move of the price), and the riskfree interest rate is 10%
per annum, compounded annually.

Solution: The terminal date payoffs are as follows: Cuu = 70, Cud = 0, Cdd = 0
[2 points]. The actual probabilities are irrelevant. The risk-neutral probabilities are
given by [3 points]:
50 = [q0 × 60 + (1 − q0 ) × 40]/1.1) → q0 = 3/4,
60 = [qu × 70 + (1 − qu ) × 50]/1.1) → qu = 4/5.
Note that the risk-neutral probabilities are time-varying and state-dependent (we don’t
need the risk-neutral probability qd because the option has 0 value at that node). We
work backwards through the tree to get [5 points]:
Cu = [qu Cuu + (1 − qu )Cud ]/1.1 = 50.91,
Cd = 0,
C0 = [q0 Cu + (1 − q0 )Cd ]/1.1 = 34.71.

(vi) (3 points) Suppose that the interest rate is constant. Show that the no-arbitrage
forward price is the risk-neutral expectation of the price of the underlying at maturity,
i.e.,
Ft = EQt (ST ),

where Ft is the forward price, ST is the price of the underlying at maturity, and Q is
the risk-neutral probability. Do not use the no-arbitrage formula of the forward price
derived in the lecture.

c LSE FM423 Page 9 of 10
 Solution: A forward position entered into at time t entails a zero payoff at t and the
payoff ST − Ft at maturity. By the risk-neutral valuation principle,

e−r(T −t) EQ
t (ST − Ft ) = 0
EQ
t (ST − Ft ) = 0

from which it follows that Ft = EQ

t (ST ) [5 points].

(vii) (7 points) A US bank’s position in options on the dollar/euro exchange rate has a delta
of 30,000 and a gamma of -80,000.

(a) Explain how these numbers can be interpreted. Are they consistent with a position
in a plain vanilla option?
(b) The exchange rate (dollars per euro) is 0.90. What position should the US bank
take in the spot foreign exchange market to make the overall position delta-
neutral?
(c) After a short period of time, the euro appreciates to $0.93. Estimate the new
delta. What additional trade is necessary to keep the position delta-neutral?

Solution: From the perspective of the US bank, the underlying asset is the euro and
the underlying price is the exchange rate expressed as the price of 1 euro in dollars
(i.e., the dollar price of the underlying asset).

(a) The delta of the current position is 30,000: this means that if the euro goes up
by 1 cent (i.e., $0.01), the value of the bank’s position increases by approximately
$300. Gamma captures (i) the second-order sensitivity of the value of the bank’s
position to changes in the price of the euro, and (ii) the first-order sensitivity of
delta to movements in the euro exchange rate. A gamma of -80,000 indicates that
when the euro goes up by 1 cent, delta decreases by approximately 800. These
numbers are consistent with a short position in a plain vanilla put (a put has a
negative delta and positive gamma) [5 points].
(b) Delta is the number of units of the underlying that you need to sell in order to
hedge the option position. Therefore, delta-hedging requires you to short 30,000
euros. The exchange rate is not needed for this calculation since we already know
the value of delta [5 points].
(c) The change in delta is approximately −80000 × 0.03 = −2400. The new delta is
30000 − 2400 = 27600. In order to remain delta-neutral, you need to buy 2400
euros, which means partially unwinding your short position in euros [5 points].

c LSE FM423 Page 10 of 10