Different Methods of Demand Forecasting | Survey of Buyer's Intentions | Delphi Method | Time Series Analysis and Trend Projection | Market Studies and Experimentation | Regression Analysis

Demand forecasting is not an easy task. Two dangers must be guarded against. First, too much emphasis should not be placed on mathematical or statistical techniques of forecasting. Though statistical techniques are essential in clarifying relationships and providing techniques of analysis, they are not substitutes for judgment. The danger is that we may go to the opposite extreme and regard forecasting as something to be left to the judgment of the so-called experts.

Related Topic: Demand Forecasting

Commonly for pure guessing, we can use following methods:

1. Survey of Buyer’s Intentions

The most direct method of estimating demand in the short run is to ask customers what they are planning to buy for the forthcoming time period generally a year. This method is also known as public opinion surveys, the most useful when bulk of the sales is made to industrial producers. In this method, the burden of forecasting is shifted to the consumers. But it would not be wise to depend wholly on the buyers’ estimates and they should be used cautiously in the light of the sellers’ own judgment.

A number of biases may creep into the surveys. If shortages are expected, customers may tend to exaggerate their requirements. The customers may know what their requirements are but they may misuse or mislead or may be uncertain about the quantity they intend to purchase from a particular firm. This method is not very useful in the case of household customers for several reasons, viz. irregularity in customers’ buying intentions, their inability to foresee their choice when faced with multiple alternatives, and the possibility that the buyers’ plans may not be real only wishful thinking.

2. Delphi Method

A variant of the opinion poll and survey method is Delphi method. It consists of an attempt to arrive at a consensus in an uncertain area by questioning a group of experts repeatedly until the responses appear to coverage along a single line or the issues causing disagreements are clearly defined. The participants are supplied the responses to previous questions from others in the group by a coordinator or leader of same sort. The leader provides each expert with the responses of the others including their reason. Others given the opportunity to react to the information or considerations advance each expert but interchange is anonymous so as to avoid or reduce halo effect and ego involvement associated with publicly expressed opinions.

It has some exclusive advantages such as: (a) It facilitates the maintenance of anonymity of the respondent’s identity throughout the course. This enables the respondent to be candid and forth right in his view. (b) Delphi renders if possible to pose the problem to the experts at one time and have their response.

Though it posses wide knowledge and experience of the subject and have an aptitude and earnest disposition towards the participants.

3. Time Series Analysis and Trend Projection

The time series relating to sales represent the past pattern of effective demand for a particular product. Such data can be presented either in a tabular form or graphically for further analysis. The most popular method of analysis of time series is to project the trend of the time-series. A trend line can be fitted through a series either visually or by means of statistical techniques such as method of least squares.

The analyst chooses a plausible algebraic relation between sales and the independent variable such as time. The trend line is then projected into the future by extrapolation. It is popular method because it is simple and inexpensive and partly because time series data often exhibit a persistent so long as the time shows a persistent tendency to move in the some direction.

4. Market Studies and Experimentation

An alternative technique for obtaining useful information about a product’s demand function involves market experiments. The firm locates one or more markets with specific characteristics, and then varies prices, packaging, advertising, and other controllable variables in the demand function with the variations occurring either over time or between markets. One market experiment technique entails examining consumer behavior is actual markets. The firm may also be able to use census or survey data to determine how such demographic characteristics as income, family size, educational level and ethnic background affect demand.

Market experiments have many serious shortcoming, they are expensive and are therefore usually undertaken on a scale too small to allow high levels of confidence in the results. Market experiments are seldom run for sufficiently long periods to indicate the long-run effects of various price, advertising or packaging strategies. The experimenter is thus forced to examine short run data and attempt to extend it to a longer period.

Various difficulties related with the uncontrolled parts of the market experiment also reduce its value as an estimating tool. A change in economic condition during the experiment is likely to invalidate the results, especially if the experiment includes the use of several separated markets, a local strike or layoffs by a major employer in one of the market areas. There is also the danger that customers lost during the experiment as a result of price manipulations cannot be regained when the experiment ends.

Market experimentation procedure utilizes a controlled laboratory experiment where in consumers are given funds with which to shop in a simulated store. By varying prices, product packaging, displays, and other factors, the experimenter can often learn a great deal about consumer behavior. The laboratory experiment, while providing similar information as field experiments, has an advantage because of lower cost and greater control of extraneous factors.

5. Regression Analysis

Regression analysis is to specify the variables that are expected to influence demand. Product demand, measured in physical units, is the dependent variable. The list of independent variables, or those which influence demand, always includes the price of the product and simply includes such factors as the prices of complementary and competitive products, advertising expenditures, consumer income and population of the consuming group. Demand function for expensive durable goods such as the houses and automobiles, include interest rates and other credit terms, those for beverages, or air conditioners include weather conditions. Demand determinants for capital goods, such as industrial machinery, include corporate profitability output to capacity ratios and wage rate trends.

Regression analysis is to obtain accurate estimates of the variables, measures of price, credit terms, output, capacity ratios, advertising expenditures, incomes and so on. Obtaining estimates of these variables is not always easy especially if the study involves data for past years. Some key variables, such as consumer attitudes toward quality and their expectations about future business conditions – which are very important in demand functions for many consumer goods, may have to be obtained by survey techniques, which introduces on element of subjectivity into the data or by market or laboratory experiments, which may produce biased data.

6. Barometric Method

A barometric or indicator, forecasting is based on the observation that there are lagged relationships among many economic time series. Changes in some series appear consistently to follow changes in one or more other series. The theoretical basis for some of these lags is obvious. For example, building permits issued precede housing starts and orders for plant and equipment lead production in durable goods industries. The reason is that each of these indicators refers to plans or commitment for the activity that follows. Other barometers are not also directly related to the economic variables they forecast. An index of common stock prices, for example, is a good leading indicator of general business activity. Although the causal relationship here is not readily apparent, stock prices reflect an aggregation of profit expectation by business managers and others and hence composite expectation of the level of business activity.

Theoretically, barometric forecasting requires the isolation of an economic time series that consistently leads the series being forecast. This relationship established; forecasting directional changes in the lagged series is simply a matter of keeping track of movement in the leading indicator. Several problems prevent such as easy solution to the forecasting problem.

• Few series always correctly indicate changes in another economic variable. Even the best leading indicators of general business conditions forecast with only to go present accuracy.
• Second, even the indicators that have good records of forecasting directional changes generally fail to lead by a consistent period. If a series is to be an adequate barometer, it not only must indicate directional changes but also, additionally, must provide a constant lead-time. Few series meet the test of lead-time consistency.
• Finally, barometric forecasting refers that even when leading indicators proved to be satisfactory from the stand point of consistently indicating directional change with a stable lead time, they provide very little information about the magnitude of change in the forecast variable.

Mainly two techniques that have been used with some success to overcome at least partially the difficulties in barometric forecasting are composite indexes and diffusion indexes. Composites indexes are weighted averages of several leading indicators. The combining of individual series into a composite index results in a series with less random fluctuation or noise. The smoother composite series has a lower tendency to produce false signals of change in the predicted variable.

Diffusion indexes are similar to composite indexes. Instead of combining a number of leading indicators into a single standardized index, the methodology consists of noting the percentage of the total number of leading indicators that are rising at given point in time.

Even with the use of composite and diffusion indexes, the barometric forecasting technique is a relatively poor tool for estimating the magnitude of change in an economic variable. Thus, although it represents a significant improvement over simple extrapolation techniques for short term forecasting.

7. Input-output Analysis

A forecasting method known as input-output analysis provides the most complete examination of all the complex interrelationships within an economic system. It shows how an increase or decrease in the demand for one’s industry output will affect other industries. An increase in the demand for trucks will lead to increased production of plastic, steel, tires, glass and other materials. The increase in the demand for these materials will have secondary effects. The increase in the demand for glass will lead to a further increase in the demand for steel, as well as for trucks used in the manufacture of glass, steel and so on. Input-output analysis traces through all these inter-industry relationships to provide information about the total input on all industries of the original increase in the demand for trucks.

It is based on set of tables that describe the interrelationships among all the component and parts of the economy. Input output analysis has a variety of uses ranging from forecasting the sales of an individual firm to probing the implications of national economic programs and policies. The major contribution of input-output analysis it that it facilitates measurement of the effects on all industrial sectors that changes in activity in any one sector.

  SOME RELATED LINKS:      

• Meaning and Concept of HRP (Human Resource Planning)

• Different methods of demand forecasting

• Environmental Scanning and Methods of Environment Scanning

• Techniques for Improving Decision Making

• Expert’s opinion survey differ from survey of Sales forces

• Steps of Demand Forecasting

• Demand Forecasting