UNIT TWO

FORECASTING
Introduction to forecasting
 A prediction, projection, or estimate of some future activity, event, or occurrence.
 Not an exact science but instead consists of a statistically tools and techniques that are supported by human
judgment and intuition.
 Business forecasting generally attempts to predict future customer demand for firms goods or services.
 Experience can lead to simple “rules of thumb,” or heuristics, that provide quick forecasts for rapid decision-
making
Elements of a Good Forecast
1. The forecast should be timely
2. The forecast should be accurate and the degree of accuracy should be stated
3. The forecast should be reliable (consistent)
4. The forecast should be expressed in meaningful terms. Different users may need different terms (Financial in dollars, marketing
in units by products, operations by units, human resources by skills, etc.)
5. The forecast should be in writing
6. The forecasting technique should be simple to understand and use
Forecasting time horizons
Would you plan for the next one year or next 10 years? Is the accuracy of forecast equal for both?
A forecast is usually classified by the future time horizon that it covers.
Time horizon falls in to three categories:
 Short-range forecast. This forecast has a time span of up to one year but is generally less than three months. It
is used for planning purchasing, job scheduling, workforce levels, job assignments, and production levels.
 Medium-range forecast. A medium-range, or intermediate, forecast generally spans from 3 months to 3 years.
It is useful in sales planning, production planning and budgeting, cash budgeting, and analyzing various
operating plans.
 Long-range forecast. Generally 3 years or more in time span, long-range forecasts are used in planning for
new products, capital expenditures, facility location or expansion, and research and development.
Steps in the forecasting process
What parameters have to be considered to forecast in your organization?
Step 1 Decide what to forecast.
Step 2 Evaluate and analyze appropriate data.
Step 3 Select and test the forecasting model.
Step 4 Generate the forecast
Step 5 Monitor forecast accuracy.
 Monitor the forecast to see if it is performing in a satisfactory manner. If it is not, reexamine the method,
assumptions, validity of data, and so on, modify as needed, and prepare a revised forecast.
 Methods of forecasting
Is your organization’s forecasting practice start from top management or lower level management? Which practice
is best?
Forecasting is some time done by a top-down method. In other cases, the reverse, a bottom-up method is used. And
in still other cases, past experience is extrapolated into the future by using mathematical and statistical procedures
Approaches to forecasting:-
 Would you use your experience or use some mathematical manipulation in your forecasting? Which one is
best?
 There are two general approaches to forecasting, qualitative and quantitative.
 Quantitative forecasts use a variety of mathematical models that relay on historical data and/or causal
variables to forecast demand.
 Qualitative forecasts incorporate such variables as the decision maker’s intuition, emotions, personal experiences, and value
system in reaching a forecast. These factors are often omitted or down played when quantitative techniques are used
because they are difficult or impossible to quantify.
 Actually, companies use both top-down and bottom-up methods at the same time and combine the resulting
projections into a single forecast. But before settling on a final forecast, they may also use the “jury of
executive opinion” (Delphi) approach to adjust judgmentally the more technically determined forecasts.
 They use most common method is simply to extrapolate past demands for each item into the future by
mathematical and statistical procedures.
Approaches to forecasting:-
 Would you use your experience or use some mathematical manipulation in your forecasting? Which one is
best?
Qualitative forecasts
It incorporate such variables as the decision maker’s intuition, emotions, personal experiences, and value system in reaching a
forecast. These factors are often omitted or down played when quantitative techniques are used because they are difficult or
impossible to quantify.
Overview of qualitative forecasting
 What are the methods that you use in qualitative forecasting?
 Qualitative techniques are subjective or judgmental in nature and are based on estimates and opinions. When a
forecast must be prepared quickly, there is usually not enough time to gather and analyze quantitative data.
 Qualitative techniques are subjective or judgmental in nature and are based on estimates and opinions. When a forecast
must be prepared quickly, there is usually not enough time to gather and analyze quantitative data.
 At other times, especially when political and economic conditions are changing, available data may be
obsolete, and more-up-to date information might not yet be available.
 What is the difference between qualitative and quantitative forecast?
 Executive opinions:
A small group of upper level managers (E.g. marketing, engineering, manufacturing, and financial managers) may
meet and collectively develop a forecast.
 This approach is often used as a part of long rang planning and new product development.
 It has the advantage of bringing together the considerable knowledge and talents of these top management people.
 Sales force composite:
The sales staff is often a good source of information because of its direct contact with consumers. Thus, these people are
often aware of any plans the customers may be considering for the future.
 What is the difference of between executive opinions and sales force composite methods of forecasting?
Consumer Surveys:
 Since the customer is the one who will ultimately determine demand, it would seem natural to solicit input
from customers
Opinions of managers and staff:
 A manager may use a staff to generate a forecast or to provide several forecasting alternatives from which to
choose. At other times, a manager may solicit opinions from a number of other managers and/or staff people
Overview of quantitative methods
 What are the methods that you use in quantitative forecasting?
 Quantitative forecasting methods use historical data.
 Quantitative methods fall in to two categories: Time series models and associative/causal models.
 Time serious models
 A time series is a time-ordered sequence of observations taken at regular intervals over a period of time (e.g. hourly, daily,
weekly, monthly, quarterly, and annually).
 The data may be measurements of demand, earnings, profits, shipments, acquirements, output etc. Forecasting
techniques based on time series data are made on the assumption that future values of the series can be
estimated from past values.
 The time is independent variable because it doesn’t dependent on the other variable. The other variable
becomes a dependent variable when it depends on time
 Time serious models include the following techniques to forecast:
 1. Naive forecasts
2. Moving averages
3. Exponential smoothing
4. Trend projection
1) Naive forecasts
 The simplest way to forecast is too assume that demand in the next period will be equal to demand in the most
recent period.
 There are two problems with this approach. First, it requires keeping all of the past data, no matter how long
the time series records are kept. Second, it will become weighed too much toward in the distant past (be too
smooth) and fail to be responsive to any changes in recent data.
 Casual method use a very different logic to generate a forecast. They assume that the variable we wish to
forecast is somehow related to other variables in the environment.
 Uses explanatory variables to predict the future.—Regression Analysis.
 A naive forecast for any period equals the previous period’s actual value
 Low cost, easy to prepare, easy to understand, but less accurate forecasts
 Can be applied to seasonal or trend data
Examples:
 If last week’s demand was 50 units, the naive forecast
 for the coming week is 50 units.

 If seasonal pattern exists, the naive forecast for next January

 would equal the actual demand for January of this year.
Example
 A restaurant is forecasting sales of chicken dinners for the month of April. Total sales of chicken dinners for
March were 320. If management uses the naïve method to forecast, what is their forecast of chicken dinners for
the month of April?
Simple Mean or Average
 One of the simplest averaging models is the simple mean or average. Here the forecast is made by simply
taking an average of all data:
Example
 New Tools Corporation is forecasting sales for its classic product, Handy-Wrench. Handy Wrench sales have
been steady, and the company uses a simple mean to forecast. Weekly sales over the past five weeks are
available. Use the mean to make a forecast for week 6.
Simple Moving Average
 The simple moving average (SMA) is similar to the simple average except that we are not taking an average of
all the data, but are including only n of the most recent periods in the average.
 As new data become available, the oldest are dropped; the number of observations used to compute the average is kept
constant. In this manner, the simple moving average “moves” through time
 Like the simple mean, this model is good only for forecasting level data. The formula is as follows
 Sales forecasts for a product are made using a three-period moving average. Given the following sales figures
for January, February, and March, makes a forecast for April.
 Example
 Using data from in the above Example and make forecasts for the months of June, July, August, and
September using a five-period moving average.
 Weighted Moving Average
In the simple moving average each observation is we
For example, in a three-period moving average each
observation is weighted one-third.ighted equally.
 In a five-period moving average each observation is weighted one-fifth. Sometimes a manager wants to use a
moving average but gives higher or lower weights to some observations based on knowledge of the industry
 Exponential Smoothing
Uses Moving average ,weighted and exponential.
The exponential smoothing model is a forecasting model that uses a sophisticated weighted average procedure to
obtain a forecast
 Even though it is sophisticated in the way it works, it is easy to use and understand. To make a forecast for the
next time period, you need three pieces of information:
 The current period’s forecast,
 The current period’s actual value
 The value of a smoothing coefficient, , which varies between 0 and 1
 The equation for the forecast is quite simple
 Example. The Hot Tamale Mexican restaurant uses exponential smoothing to forecast monthly usage of tabasco sauce. Its
forecast for September was 200 bottles, whereas actual usage in September was 300 bottles. If the restaurant’s managers use
an of 0.70, what is their forecast for October?
Techniques for Trend
 Analysis of trend involves developing an equation that will suitably describe trend (assuming that trend is
present in the data). The trend component may be linear, or it may not.
 A simple plot of the data often can reveal the existence and nature of a trend. The discussion here focuses exclusively on
linear trends because these are fairly common.
 There are two important techniques that can be used to develop forecasts when trend is present. One involves use of a trend
equation; the other is an extension of exponential smoothing.
 Trend Equation
 For example, consider the trend equation F t =45 + 5t. The value of F when t = 0 is 45, and the slope of the line is
5, which means that, on the average, the value of Ft will increase by five units for each time period. If t = 10, the
forecast, Ft , is 45 + 5(10) = 95 units. The equation can be plotted by finding two points on the line.
One can be found by substituting some value of t into the equation (e.g., t= 10) and then solving for F t . The other point is a (i.e., Ft
at t= 0). Plotting those two points and drawing a line through them yields a graph of the linear trend line. The coefficients of the line, a
and b, are based on the following two equations:
Example
 Mike Clem, owner of Clem’s Competition Clutches, designs and manufactures heavy-duty car clutches for use
in drag racing. In his first 10 months of business, Mike has experienced the demand shown in Table below.
 Using the month as the independent variable (x) to forecast demand (y), Mike wants to develop a linear
regression forecasting model and use the model to forecast demand for months 11,12,and 13.Following
Equations stated above the first step is to set up columns to calculate the average x and y values, as well as the
sums of the x, y, x2 and xy values for the first 10 months:
Casual Model
Recall that causal, or associative, models assume that the variable we are trying to forecast is somehow related to other variables in
the environment.
 Some of the best-known causal models are
regression models. In this section we look at
linear and multiple regression and how they are
used in forecasting.
 Linear Regression
In linear regression the variable being forecasted,
called the dependent variable, is related to some other
variable, called the independent variable, in a linear (or
straight-line) way. Figure 5 shows how a linear
regression line relates to the data.
 The relationship between two variables is the
equation of a straight line:
 The steps in computing the linear regression
equation are as follows:
 A maker of personalized golf shirts has been
tracking the relationship between sales and
advertising dollars over the past four years.
The results are as follows:
Measuring forecast Accuracy
 One of the basic principles of forecasting is that forecasts are rarely perfect
 One of the most important criteria for choosing a forecasting model is its accuracy. Also, data can change over
time, and a model that once provided good results may no longer be adequate
 However, error for one time period does not tell us very much. We need to measure forecast accuracy over
time. Two of the most commonly used error measures are the mean absolute deviation (MAD) and the mean
squared error (MSE). MAD is the average of the sum of the absolute errors:
 Forecast error= Actual- forecast
 MAD=Summation of AV forecast error/n
 MSE=summation of Error 2/n
 MAPE=|Error|/Actual *100
 %Error=AV/At*100
Week Sales 3MA Error AV Error 2 %Error

1 39
2 44
3 40
4 45 41 4 4 16 8.89%(4
/45)*100
5 38 43 -5 5 25 13.16%
6 43 41 2 2 4 4.65%
7 39 42 -3 3 9 7.69%
8 40 14/4=3. 54/4=13 34.39%/
5(MAD) .5(MSE) 4=8.6%
(MAPE)
The End