11
1 DEMAND FORECASTING DEMAND FORECASTING 3 W’s: Weather, Work and Women are 3 W’s: Weather, Work and Women are unpredictable unpredictable In modern business, production is often made in anticipation of demand. Anticipation of demand implies “Demand “Demand Forecasting”. Forecasting”. It means expectation about the future course of development which is uncertain, but not entirely so. Hence, one can hopefully predict the future events and reasonably gain. In particular, the demand forecasting means expectation about the future course of the market demand for a product. It based on the statistical data about past behavior and empirical relationships of the demand determinants.

Demand Forecasting

Embed Size (px)

Citation preview

Page 1: Demand Forecasting

1

DEMAND FORECASTINGDEMAND FORECASTING

3 W’s: Weather, Work and Women are 3 W’s: Weather, Work and Women are unpredictableunpredictable

In modern business, production is often made in anticipation of demand.Anticipation of demand implies “Demand “Demand Forecasting”.Forecasting”.It means expectation about the future course of development which is uncertain, but not entirely so.Hence, one can hopefully predict the future events and reasonably gain.In particular, the demand forecasting means expectation about the future course of the market demand for a product.It based on the statistical data about past behavior and empirical relationships of the demand determinants.

Page 2: Demand Forecasting

2

Demand forecasting may be undertaken at the following levels:

1. Micro level: it refers to the demand forecasting by the individual business firm for estimating the demand for its product.

2. Industry Level: it refers to the demand estimate for the product of the industry as a whole. It is under taken by an industrial or trade association. It relates to the market demand as a whole.

3. Macro Level: it refers to the aggregate demand for the industrial output by the nation a whole.

It is based on the National income or the aggregate expenditure of the country.

Country’s consumption function provides an estimate for the demand forecasting at macro level.

With the growth of national income, consumption expenditure increases, as such, overall demand for goods and services in general may tend to rise.

Page 3: Demand Forecasting

3

Methods of Demand Forecasting:Methods of Demand Forecasting:1.1. Sources of Data Collection for Demand Forecasting:Sources of Data Collection for Demand Forecasting:(a)(a) Primary Data: Primary data or information are original in character Primary Data: Primary data or information are original in character

which are collected for the first time for the purpose of analysis. which are collected for the first time for the purpose of analysis. Primary data are raw data and require statistical processing.Primary data are raw data and require statistical processing.

Sources of Primary data:Sources of Primary data: (i). Market Survey or Survey Method(i). Market Survey or Survey Method: the market survey is the most direct : the market survey is the most direct

approach to demand forecasting in short run. It may be conducted approach to demand forecasting in short run. It may be conducted through Consumer survey and collective Opinions.through Consumer survey and collective Opinions.

Consumer Survey:Consumer Survey: A sample survey of the consumer may be undertaken questioning them A sample survey of the consumer may be undertaken questioning them

about what they are planning or intending to buy.about what they are planning or intending to buy. A questionnaire may be prepared in this regard.A questionnaire may be prepared in this regard. This may be mailed to the consumer or can be sent through the field This may be mailed to the consumer or can be sent through the field

investigator.investigator. The informations collected through questionnaire are to be thoroughly The informations collected through questionnaire are to be thoroughly

scrutinized and edited to check the inconsistencies, inaccuracy and scrutinized and edited to check the inconsistencies, inaccuracy and incorrectness of the data supplied.incorrectness of the data supplied.

The data may be classified and tabulated for systematic presentation The data may be classified and tabulated for systematic presentation and analysis.and analysis.

Page 4: Demand Forecasting

4

Sometime, personal observations or the personal interviews method may also be adopted to collect informations when there is a small sample size of consumer survey.

On the basis of data/informations collected in the consumer survey, the managerial economist/manager construct important demand relationships, such as, price-demand, income-demand, advertisement-demand, cross demand etc.

Drawbacks:1. Expensive2. Time Consuming3. It is likely that the consumers may not be able to give

correct answers and may not also cooperate in the process.

4. Informations may not be complete.5. Cheating on the part of enumerators may also vitiate the

result of the survey.

Page 5: Demand Forecasting

5

Through Collective Opinion:Through Collective Opinion: Under this method, the sales man have to report to the head

office with their estimates on expectations of sales in their territories.

Such informations cal be obtained from the retailers and whole sellers by the company.

By aggregating these forecasts a generalization on an average is made, which is also based on the value judgment and collective wisdom of the top level sales executive, marketing manager, business/managerial economist all together.

The opinion method is cheaper and easy to handle. It is less time consuming also.

Its main draw back, however, is that it is subjective and subject to the high element of bias of the reporting agency.

It is also called “hunch” method of forecasting as it is based on the hunches of the high level experts.

It can not be relied on the long term business planning.

Page 6: Demand Forecasting

6

(ii) Market Experimentation:(ii) Market Experimentation:There may be two types of market experimentation such

as (a) Experiments in Laboratory and (b) Test marketing.Experiment in Laboratory:Experiment in Laboratory: In this method, a consumer clinic or small laboratory is

formed by creating an artificial market situation. To study consumers reaction to a change in demand

variables, in the controlled conditions of the consumer clinic, the selected consumers are given a small amount of money and asked to buy certain item.

The consumer behaviour in the clinic is thus observed and inferences are drawn.

Drawbacks:o It is expensive and time consuming method.o The stimulated market situation may not be fully

representing the actual market situation.

Page 7: Demand Forecasting

7

Test Marketing: In this method, a a market experiment is performed under actual

market conditions. First, a choice of the market for experiment is made and is

segregated from the rest. The business firm then conducts the experiments in this market,

under controlled conditions, by varying one or more of demand determinants like price, advertisement, package etc.

Consumer behaviour in this market is then observed and recorded. This may help in assessing the effect of change in the determining

variables on the consumer demands for the product. Unlike consumer survey, in case market test, demand is estimated

on the basis of actual purchases (and not intentions and expectations) of the consumer under the given partially known conditions.

Drawbacks:o Very expensiveo It is difficult to plan market experimentation under heterogeneous

market conditions.o It is often conducted in developed countries like USA, not in India

so far.

Page 8: Demand Forecasting

8

(b) Secondary Sources of Data: There are ample secondary sources for collecting the required

information for market and demand analysis. The main source of such data are:

(i) Official Publication of the central, state and local govt. such planned documents, census of India, Statistical abstracts of the Indian union, annual survey of industries, annual bulletin of statistics of exports and imports, monthly studies of productions of selected industries, economic survey, National Sample Survey reports etc.

(ii) Trade and technical or economic journals and publications like the EPW, Indian Economic journal, stock exchange directory, basic statistics and other informations supplied by the centre for monitoring Indian economy etc.

(iii) Official publications like the reserve bank of India, i.e. annual report and currency and finance, monthly bulletin of reserve bank of India etc.

(iv) Publications brought out by research institutions, universities, associations etc.

(v) Unpublished data such as firms account books showing data about sales, profits etc.

Page 9: Demand Forecasting

9

2. 2. Statistical Methods of Forecasting Statistical Methods of Forecasting Demand:Demand:

Once market demand data are collected by the market survey or from sales record of the farm demand forecasting can be possible from such information.

In statistical analysis, economists usually resort to two types of data for demand estimation, 1. time series data, 2. Cross section data and 3. Pooled or panel data.

Time Series data: time series data refer to data collected over a period of time recording historical changes in price, income and other relevant variables influencing demand for a commodity.

Time series analysis relates to the determination of change in a variable in relation to time.

Usually trend projections are important in this regard.

Page 10: Demand Forecasting

10

Cross-Section Data: cross-sectional analysis is undertaken to determine the effects of changes in determining variables like price, income etc. on the demand for a commodity, at a point of time.

In time series analysis for instance, for measuring income elasticity of demand, a sales income relationship may be established from the historical data and their past variations.

In cross-sectional analysis, however, different levels of sales among different income groups may be compared at a specific point of time.

The income elasticity is then estimated on the basis of differences so measured.

Pooled or Panel DataPooled or Panel Data: It implies a combination and time series and cross sectional data.

Page 11: Demand Forecasting

11

Criteria of a Good Forecasting Method:Criteria of a Good Forecasting Method:Joel Dean (1976)Joel Dean (1976) lays down the following criteria of a lays down the following criteria of a

good forecasting method:good forecasting method: Accuracy:Accuracy: Forecast should be accurate as far as possible. Forecast should be accurate as far as possible.

Its accuracy must be judged by examining the past forecasts Its accuracy must be judged by examining the past forecasts in the light of the present situations.in the light of the present situations.

Plausibility:Plausibility: it implies managements understanding of the it implies managements understanding of the method used for forecasting. It is essential for a correct method used for forecasting. It is essential for a correct interpretation of the results.interpretation of the results.

Simplicity:Simplicity: A simpler method is always more comprehensive A simpler method is always more comprehensive than a complicated one.than a complicated one.

Economy:Economy: it should involve lesser costs as far as possible. it should involve lesser costs as far as possible. Its costs must be compared with the benefits of forecasting.Its costs must be compared with the benefits of forecasting.

Quickness:Quickness: It should yield quick results. A time consuming It should yield quick results. A time consuming method may delay the decision making process.method may delay the decision making process.

Flexibility:Flexibility: Not only the forecast has to maintained up to Not only the forecast has to maintained up to date, there should be possibility of changes should be date, there should be possibility of changes should be incorporated in the relationship entitled in forecast procedure, incorporated in the relationship entitled in forecast procedure, time to time.time to time.