Demand Estimation and

Forecasting

Lecture 2
 What is Demand Estimation?

• The science and art of converting the

qualitative understandings of a market into
quantitative data.
• But why?
• Uncertainty in business activities
• Risk involved in business activities
• Success depends on estimation &
forecasting
 Objectives of Demand Estimation
• How much to sell:
– Fixing the product/ service input targets.
• Where to sell:
– Geographical area -Identification of market area i.e. provinces,
districts, group of villages etc
– Product range - determination of inputs based on desired outputs
– Profile of customers - large, commercial or small scale producers
• When to sell:
– To determine the period of peak, low and no demand
– To determine how inputs will be used
– To determine if inputs are needed for general use or for special
requirements
 Tools or Techniques of Demand
Estimation
Consumer • Questioning the consumer to obtain
surveys relevant information

Consumer clinic • Experimental group to understand

or focus group relationship between variables

• Direct market experiments to

Market understand the changes in demand due
experiments to changes in dependent variables

Statistical • The regression analysis

techniques
 Importance of demand
estimation
• Input demand estimates help producers in:
– Preparation of business plans
– Making the financial arrangements
– Working out quantity of inputs to procure
– Arranging the storage & transportation
– Formulation of business strategy
– Developing the sales promotion strategy
 Process and Steps involved in Demand
Estimation
• Identification of a variable: Dependent &
Independent
• Development of model: mathematical equation
e.g. Q=α1 – α2Px + α3Y + u
• Collection of data: Primary and Secondary
• Estimation of parameter of the model
• Development of estimates based on the model
 Regression Analysis
• What Is a Statistical Relation?
– A statistical relation exists when averages are related.
– A deterministic relation is true by definition.
• Specifying the Regression Model
– Dependent variable Y is caused by X.
– X variables are independently determined from Y.
• Least Squares Method
– Minimize sum of squared residuals.
 How to Improve Accuracy of DE

• Define the objectives clearly

• Decide the level of accuracy required
• Select an appropriate method
• Select reliable sources of data
• Depute competent and experienced
persons to collect and analyze data
 Limitations of Demand Estimation

• The model cannot be exact

• Fluctuations in quantitative and qualitative
nature of consumer behavior
• Indifferent consumer behavior
Demand Forecasting
 What is Demand Forecasting?

• Demand forecasting is a scientific and

analytical estimation of demand for a
product/service for a specified period of
time
– What will be sale of gold in the next three
years…
– Long-run forecasting cover 5,10 or 20 years
– Short-run cover 3 months or 6 months
 Objectives of Demand Forecasting

• To prepare suitable sales and production policy

• To ensure the purchase of raw material and minimize cost
• Setting sale target and utilization of maximum resources
• To determine appropriate advertising policy and
promotion program
• Appropriate production scheduling to avoid over
production
• It help to determine suitable price policy & cost strategy
• To determine financial requirement
Categorization by level of Forecasting:

Firm • Forecasting by individual firm products

• Important category for managerial decision

level making – marketing & production

Industry • Forecasting of product in an industry as a

whole

level • Insights to industrial growth pattern

Macro • Forecasting of aggregate demand in the

economy
• Helps in policy formulation at government
level level
Categorization by Time Period:
• Usually for a period less than a
year
Short • Avoid underproduction and
overproduction
term • Helps achieve Sales target &
appropriate pricing

• Time horizon of 5 – 7 years, 10 –

20 years
Long • Helps in manpower planning,
long term capital requirement,
term investment decision,
interdependence of industry
 Steps involved in demand forecasting

• The following are the typical steps for a

systematic demand forecasting:
– Understanding the objective
– Determining the time perspective
– Understand and identify customer segments
– Identify major factors that influence demand forecast
– Determining the appropriate forecasting technique
– Estimation and interpretation of results
Methods of Demand Forecasting

• Consumer Survey
Survey • Expert opinion
Method • Consumer clinic

• Trend projection method

Statistical • The regression and correlation
method
techniques • Econometric methods
 Time-Series Forecasts
• A time-series model shows how a time-
ordered sequence of observations on a
variable is generated
• Simplest form is linear trend forecasting
– Sales in each time period (Qt ) are assumed to
be linearly related to time (t)

Qt  a  bt
 Linear Trend Forecasting
• Use regression analysis to estimate
values of a and b
Qˆ t  aˆ  bt
ˆ
• If b > 0, sales are increasing over time
• If b < 0, sales are decreasing over time
• If b = 0, sales are constant over time

• Statistical significance of a trend is

determined by testing b̂ or by examining
the p-value for b̂
Linear Regression Forecasting

• Identify dependent (y) and

independent (x) variables
• Solve for the slope of the line

b
 XY  n XY

 X  nX2 2

• Solve for the y intercept

a  Y  bX
• Develop your equation for the
trend line
Y=a + bX
 Linear Regression Problem: A maker of golf shirts has been tracking
the relationship between sales and advertising dollars. Use linear
regression to find out what sales might be if the company invested
$53,000 in advertising next year.

Sales $ Adv.$ XY X^2 Y^2 b

 XY  n XY

 X  nX
2
(Y) (X) 2

1 130 32 4160 2304 16,900

28202  447.25 147.25 
2 151 52 7852 2704 22,801 b   1.15
9253  447.25 
2

3 150 50 7500 2500 22,500 a  Y  b X  147.25  1.1547.25 

a  92.9
4 158 55 8690 3025 24964
Y  a  bX  92.9  1.15X
5 153.85 53 Y  92.9  1.1553   153.85

Tot 589 189 28202 9253 87165

Avg 147.25 47.25
Seasonal (or Cyclical) Variation
• Can bias the estimation of parameters in
linear trend forecasting
• To account for such variation, dummy
variables are added to the trend equation
– Shift trend line up or down depending on the
particular seasonal pattern
– Significance of seasonal behavior determined by
using t-test or p-value for the estimated
coefficient on the dummy variable
Sales with Seasonal Variation

 
 

     





2004 2005 2006 2007

 Smoothing Techniques
• Series do not show continues trend – there
are always seasonal and random variations

• This technique is used to smoothen these

variations and then forecasting of future
values.
 Smoothing Techniques…
• Moving average: forecasts on the basis of
demand values

• Weighted Moving average : forecasts on the

basis of weights of the recent observations.

• Exponential smoothing: assign greater weight to

most recent data as to have realistic estimate of
the fluctuations.
 Exponential Smoothing
• Most widely practiced method of time series
forecasting
• Weighted average of two variables
Ft 1  αA t  1  α Ft
Where….
F = forecast for next period
t +1

A = actual value for present period

t

F = previously determined forecast for present period

t

α = weighting factor (between 0 and 1)

 Adjusted Exponential Smoothing
 A method that uses measurable, historical data
observations to make forecasts by calculating the
weighted average of the current period’s actual
value and forecast, with a trend adjustment
added in.
 When to use the method
 Preferred Scenario:
 When a trend is present
 Good Scenario:
 When there is a cyclical or seasonal pattern
 Adjusted Exponential Smoothing
• To use the method, three steps are involved:

 Step 1 - Smoothing the level of the series

S t  αA t  (1  α)(S t 1  Tt 1 )
 Step 2 – Smoothing the trend

Tt  β(S t  S t 1 )  (1  β)Tt 1
 Forecast including the trend
Where:
FITt 1  S t  Tt  Tt = trend factor for current period
 Tt+1 = trend factor for next period
 β = smoothing constant
Example: Coca-cola company uses exponential smoothing with trend to
forecast its sales turnover. At the end of July the company wishes to forecast
sales for August. July demand was 62. The trend through June has been 15
additional crates of soda sold per month. Average sales have been 57 crates per
month. The company uses alpha+0.2 and beta +0.10. Forecast for August.

• Smooth the level of the series:

S July  αA t  (1  α)(S t 1  Tt 1 )  0.2 62   0.8 57  15   70

• Smooth the trend:
TJuly  β(St  St 1 )  (1  β)Tt 1  0.170  57   0.9 15  14.8

• Forecast including trend:

FIT August = St + Tt = 70 + 14.8 = 84.8 Crates

 Limitation of Demand forecasting

• Past data and events are not always true

predictors of future
• If the forecaster is inefficient or inexperienced,
the result may be wrong
• Demand depends on psychology of consumer
• It is only an estimation, a certain amount of error
is always there