2. Chapter Outline
Meaning of Demand Forecasting
Techniques of Demand Forecasting
Subjective Methods of Demand Forecasting
Quantitative Methods of Demand Forecasting
Limitations of Demand Forecasting
3. Demand forecasting
Demand forecasting is the scientific and
analytical estimation of demand for a
product (service) for a particular period of
time.
It is the process of determining how much
of what products is needed when and
where.
4. Categories of forecasting
Macro level
GDP
Components of GDP
Share of manufacturing in total GDP in 2011
Industry level
Industry sales- car sales in 2011
Firm level
Sales of Tata Indica in 2011
5. Choice of forecasting technique
Objective of forecast
New product, impact of advertisement
Cost effective
Opportunity cost of resources employed
Time perspective
Urgency of forecast- breaking of epidemic
Long run/short run
Availability of data
Quality and quantity
6. Techniques of Demand Forecasting
Qualitative (subjective) technique
Rely on human judgment and opinion
Experts’ Opinion
Group Discussion
Delphi Method
Sales force composite
Opinion polls
Market research
Market simulation
Test marketing
Surveys of spending plans
Barometric technique
7. Quantitative technique
Use mathematical or simulation models
Based on historical demand data or
Relationship between demand and other
variables
Naïve techniques
Trend Projections
Smoothing Techniques
Econometric Techniques
Techniques of Demand Forecasting
8. Qualitative Technique
Experts opinion
Group Discussion
Within a corporation (jury of executive opinion)
Structured discussion on topics/ forums
Eg., stock market, beauty products
9. Experts opinion
Delphi Method (Rand Corporation in 1950)
Forecast the impact of technology on warfare
Experts do not meet face to face
Sequential series of written Q & A
Consolidated opinions of experts is sent for
revised views till conclusions converge on a
point.
Qualitative Technique
10. Expert’s opinion- Delphi Method
Merits
Decisions are enriched with the experience of
competent experts.
Very useful when product is absolutely new to all
the markets.
Demerits
Experts’ may involve some amount of bias.
Sometimes difficulty in assessing the degree of
expertise
With external experts, risk of loss of confidential
information to rival firms.
Qualitative Technique
11. Sales force composite
Salespersons are asked about estimated sales
targets in their respective sales territories in a
given period of time.
Merits
Cost effective as no additional cost is incurred
on collection of data.
Estimated figures are more reliable, as they
are based on the notions of salespersons in
direct contact with their customers.
Qualitative Technique
12. Demerits
Results may be conditioned by the bias of
optimism (or pessimism) of salespersons.
Salespersons may be unaware of the economic
environment of the business and may make
wrong estimates.
This method is ideal for short term and not for
long term forecasting
Qualitative Technique
13. Opinion poll (Buyers’ opinion, consumers
opinion survey)
consumers future buying intentions of
Products
Brand preferences
Quantities purchased
Response to price increase
Implied comparison with competitor’s products
Census Method: Involves contacting each and every
buyer
Sample Method: Involves only representative sample
of buyers
Qualitative Technique
14. Opinion Poll
Merits
Simple to administer and comprehend
Suitable when no past data available
Suitable for short term decisions regarding
product and promotion
Demerits
Expensive both in terms of resources and time
Investigators’ bias regarding choice of sample
and questions
Qualitative Technique
15. Market research
Market simulation
create “artificial market”, consumers are
instructed to shop with some money.
“Laboratory experiment” ascertains consumers’
reactions to changes in price, packaging, and
even location of the product in the shop
Grabor-Granger test for pricing strategy
Qualitative Technique
16. Market Research- Market simulation
Merit
Provides information on changing consumer
behaviour and impact of determinants of
demand
Very useful in case of new products
Demerits
People behave differently when they are being
observed.
In Grabor-Granger tests consumers may not
quote the price they may pay
Qualitative Technique
17. Market Research
Test marketing
product is actually sold in certain segments of the
markets
Location, no. of test markets, duration of test are
very crucial to the success of the results.
Merits
Most reliable among qualitative methods.
Very suitable for new products.
Less risky than launching the product directly
Qualitative Technique
18. Market Research- Test marketing
Demerits
Costly
Requires actual production
Failure means entire cost of test is sunk.
Time consuming
Extrapolation may not give accurate results
Markets are geographically widely distributed
Qualitative Technique
19. Qualitative Techniques
Surveys of spending plans
More macro type of study
Income spending habits of consumers
NSSO survey is India on consumer expenditure
Proportion of income spent on various items
20. Barometric Technique
Alert economic conditions.
Helps in predicting future trends on the basis
of index of relevant economic indicators
Particularly helpful when past data do not show
any trends
Eg., forecasting the impact of recession of
2008-09
Qualitative Techniques
21. Qualitative Techniques
Indicators may be
Leading indicators
Indicators that move ahead of economic events
Export-import values, Building permits
Coincident indicators
Move up or down simultaneously with economic
activity
Industrial production
Lagging indicators
Move with economic series after a period of time
Average duration of employment, commercial and
industrial loan outstanding
23. Quantitative Techniques
Constant compound growth rate
Appropriate when variable is expected to
increase at a constant percentage
CGR = (E/B) (1/n)
-1 => E/B = (1+i)n
E : ending value, B : beginning value, n: no. of
years, i: growth rate
Demerit
Does not take the fluctuations into
consideration
24. Quantitative Techniques
Trend Projections
Statistical tool to predict future values of a variable
on the basis of time series data
Secular Trend (T)
Direction of movement of data over long period of time
Cyclical trend (C)
Business cycles
Seasonal trend (S)
Seasonal variations within a year
Random events (R)
Have no trend of occurrence
25. Additive Form: Y = T + S + C + R
Multiplicative Form: Y = T x S x C x R
Log Y= log T + log S + log C + log R
Methods of trend projection
Graphical
Linear regression models
ARIMA or Box Jenkins Method
Quantitative Techniques
26. Smoothing Techniques
Moving Average
Based on averages of recent past data
Et+1 = ( Xt +Xt-1+…+Xt-N-1) / N
E: forecast, X: actual observation
Eg., 3 month moving average, 5 month moving
average
Weighted moving average
Attaching weights to the past data
Et+1 = ( w1Xt +w2Xt-1+…+wnXt-N-1) / N
Quantitative Techniques
27. Exponential smoothing
Greater weight is assigned to most recent data
Et+1 = wXt + (1-w)Et
0< W <1
Larger the W, greater the importance of the
observation
If series is volatile – less smoothing effect
Quantitative Techniques
29. Limitation of Demand forecasting
Changing fashion changes preferences of
consumers
Consumers psychology
Understanding of this is very difficult
Uneconomical for small firms
Time required to do the analysis
Data collection is costly
Lack of experienced experts
Lack of past data
30. Limitations of Demand Forecasting
Change in Fashion:
Consumers’ Psychology: Results of forecasting depend
largely on consumers’ psychology, understanding which
itself is difficult.
Uneconomical: Requires collection of data in huge
volumes and their analysis, which may be too expensive for
small firms to afford. Estimation process may take a lot of
time, which may not be affordable.
31. Lack of Experienced Experts: Accurate
forecasting necessitates experienced
experts, who may not be easily available.
Forecasting by less experienced individuals
may lead to erroneous estimates.
Lack of Past Data: Requires past sales
data, which may not be correctly
available. Typical problem in case for a
new product.