Techniques of Forecasting
The various techniques of forecasting demand may be grouped under the following categories:
Qualitative Techniques/Opinion Polling Method
In this method, the opinion of the customers, sales force and experts could be gathered to determine the emerging trend in the market.
The opinion polling methods of demand forecasting are of three different kinds:
1) Consumer Survey Methods: The most direct method of forecasting demand in the short run is the survey method. Surveys are conducted to collect information about future purchase plans of the probable customers of the product. Survey methods include the following:
i) Complete Enumeration Survey: Under the complete Enumeration Survey, the company has to go for a door-to-door survey for the forecast period by contacting all the households in the area.
ii) Sample Survey and Test Marketing: Under this method, some representative households are selected on a random basis as samples and their opinion is taken as the generalised opinion. This method is based on the basic assumption that the sample actually represents the population. A variant of the sample survey technique is test marketing. Product testing essentially involves placing the product with several users for a set period. Their reactions to the product are noted after some time and an estimate of likely demand is made from the result.
iii) End-Use Method: In this method, the sale of the product under consideration is projected based on a demand survey of the industries using this product and the intermediate product. In other words, demand for the last product is the end-use demand for the intermediate product used in the production of this final product.
2) Sales Force Opinion Method: This is also known as the Collective Opinion Method. In this method, instead of consumers, the opinion of the salesmen is taken. It is sometimes referred to as the “grassroots approach” as it is a bottom-up method that requires each salesperson in the company to make an individual forecast for his or her particular sales territory. These particular forecasts are discussed and agreed upon with the sales manager. The composite of all forecasts then constitutes the sales forecast for the company.
3) Delphi Technique: This method is also known as the Expert opinion method of investigation. In this method instead of depending upon the opinions of buyers and salesmen, firms can obtain the views of the specialists of experts in their respective fields. Opinions of different experts are taken and their identity is kept secret. These opinions are then exchanged among the various experts and their reactions are sought and analysed. The process goes on until some sort of unanimity is arrived at among all the specialists. This method is best suited in cases where intractable changes are occurring.
Quantitative Techniques/Statistical or Analytical Methods
The statistical methods, which are frequently used, for making demand projections are given below:
1) Trend Projection Method: An old firm can use its own data of past years regarding its sales in the past 750 years. These types of data are known as time series of sales. A dry can predict sales of its product by the fitting trend to the time series of sales. A trend line can be fitted by the graphical method or by algebraic equations. The equations method is more appropriate. The trend can be estimated by using any one of the methods given as follows :
i) Graphical Method: A trend line can be fitted through a series graphically. Old values of sales for various areas are plotted on a graph and a free-hand curve is drawn passing through as many points as possible. The direction of the free-hand curve indicates the trend. The main drawback of this method is that it may show the trend but not measure it.
ii) Least Square Method: The least square method is based on the assumption that the past rate of change of the variable under study will continue in the future. It is a mathematical procedure for fitting a line to a set of observed data points in such a way that the sum of the squared differences between the calculated and observed value is minimised. This technique is used to find a trend line which nicely fits the available data. This trend is then used to project-dependent variables in the future. This method is very popular because it is simple and cheap.
iii) Time Series Data: Time series data refer to data collected over a period 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 about time. Usually, trend projections are essential in this regard.
iv) Moving Average Method: Under this method, the moving average of the sales of the past years is computed. The computed moving average is taken as the forecast for the next year or period. This is based on the belief that future sales are the average of past sales. The moving average is the process of computing the average by leaving the oldest observation and including the next one.
v) Exponential Smoothing: Exponential smoothing is a popular technique for short-run forecasting. It uses a weighted average of past data as the base for a forecast. The process gives the heaviest weight to more recent information and a smaller weight to observations in the more distant past.
The reason for this is that the future is more dependent on the recent past than on the distant past. The method is known to be effective when there is randomness and no seasonal changes in the data.
2) Barometric Method: It is also known as leading indicators forecasting. The National Bureau of Economic Research of America has identified three types of indicators – Leading indicators, Coincidental indicators and Lagging indicators.
The analyst should establish a relationship between the sales of the product and the economic indicators to project the correct sales and to measure to what extent these indicators affect the sales. Establishing a relationship is not an easy task, especially in the case of a new product where there is no past record.
3) Regression Method: This is a very common method of forecasting demand. Under this method, a relationship is established between quantities demanded (dependent variable) and independent variables such as the price of the good, prices of the related goods, income, etc.
Once the relationship is established, a regression equation is derived assuming the relationship to be linear. The equation will be of the form Y = A + BX. There could also be a curvy linear relationship between dependent and independent variables. Once the regression equation is derived the value of Y, i.e., quantity demanded can be estimated for any given value of X.
4) Econometric Method: The econometric model forecasting involves estimating several simultaneous equations, which are, generally, behavioural equations, mathematical identities and market-clearing equations. The econometric model technique is known also as the simultaneous equations method and complete system approach to forecasting. This method uses sophisticated mathematical and statistical tools.