The sources of Data Collection for Demand ForecastingThrough a market research a variety of information .Qualitative and quantitative is called as “DATA”. This have to be collected to estimation of Demand function and demand forecasting. These information may be pertaining to varies aspects of market and demand.
Cont…. Demand in Past and Present. Nature of product. Types of consumer- Domestic consumers and Industrial consumers. Age. Sex. income of the consumers. Attitude. Preferences. Tastes. Habits.
Price Quotations in the retail and wholesale markets. Urban. Rural. Local. National. International or Global.
Sales Promotion. Advertising . Free samples. Discounts. Window display. As well as the expenditures so incurred or involved.
Primary and Secondary DataPrimary Data: Primary data or information are original in character which are collected for the first time for the purpose of analysis. Primary data are raw data and require statistical processing.
Secondary Data Secondary data or information are those which are obtained from someone else’s records. These Data are already in existence in the recorded or published forms. Secondary Data are like finished products since they have been processed statistically in some form or the other.
Secondary Sources of Data Official publications of the Central, state and local Governments. 1) Plan documents. 2) census of India. 3) Statistical Abstracts of the Indian Union. 4) Annual Survey of Industries. 5) Annual bulletin of Statistical of Exports and Imports. 6) monthly studies of production of selected Industries. 7) Economic Survey, National Sample Survey Reports.
Trade and Technical or Economic journals and publication1) Economic and political weekly.2) Indian Economic journal.3) Stock exchange directory.4) Basic statistics and other informations supplied by the centre for monitoring Indian Economy.
Official publications of International Bodies1) IMF.2) UNO.3) World Bank etc.
Market reports and trade bulletins published by stock exchange1) Trader associations.2) Large business houses.3) Chambers of Commerce, etc.
Publications brought out by research institutions1) Universities.2) Associations. etc.
Unpublished Data1) Firm’s account books 1) Sales. 2) Profits.etc. secondary data should not be taken at their face value and are never to be used blindly.
Statistical Methods of Forecasting Demand There are various methods adopted to estimate potential demand. Statistical methods are obviously more scientific, against crude value judgment used to estimate future demand. One must take a mid-way by combining statistical results with the value judgment. Again different statistical forecasting methods are not mutually exclusive. They are to be used in combination for accuracy and cross checking purposes. For forecasting purposes, it is essential to estimate the structural form and parameters of the demand function empirically.
Two types of data for demand estimation: Time series data Cross-sectional dataTime series data Time series data refer to data collected over a period of time according historical changes in price, income, and other relevant variables influencing demand for a commodity. Time series analysis relate to the determination of change in a variables in relation to time. Usually trend projections are important in this regard.
Cross-sectional 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 fast variations in cross-sectional analysis, however different levels of sales among different income groups may be compared at a specific point of time.
The Important Demand Forecasting Methods are: Consumption Level Method Trend Projection Method Regression Analysis and Econometric Method
Consumption Level Method Consumption Level demand may be estimated on the basis of the co- efficients of income elasticity and price elasticity of demand. Viewing projected income level and income elasticity of demand relationship, demand forecasting may be made as under: D*=D(1+M*.em) Where, D*=Projected per capita demand D=Per capita demand M*=Projected relative/percentage change in per capita income. em=Income elasticity of demand.
Trend Projection Method A time series analysis of sales data over a period of time is considered to serve as a good guide for sales or demand forecasting. For long-term demand forecasting trend is computed from the time based demand function data. Trends refers to the long-term persistent movement of data in one direction – upward or downward. There are two important methods used for trend projections: The method as moving averages. The least square method
The Method of Moving Averages. A moving average forecast is based on the average of a certain number of most recent periods. One can select the number of months or years or other period units in the moving average according to how for back the data is relevant to future observations.
The least square method The method of least squares is more scientific as compared to the method of moving averages. It uses the straight line equation y= a+bx, to fit the trend to the data.
Regression analysis and econometric model building Most commonly for demand forecasting proposes, the parameters of the demand function are estimated with regression analysis. In demand regression equations relevant variables have to be included with practical considerations and relevant data have to be obtained.
Examples: Personal disposal income towards demand for consumer product. Agricultural or farm incomes towards demand for the agricultural equipments, fertilizers, etc. Construction contracts for demand towards building material such as cement, bricks, steel, tiles etc. Automobile registry over a period towards demand for car spare parts, petrol etc.