What are the Methods of Demand Forecasting

Demand Forecasting is the activity in which the demand for a product or service is estimated for a future period of time. There are several methods of demand forecasting applied in terms of:-

  • Purpose of forecasting
  • Data required
  • Data availability
  • The time frame for forecasting.

Each method varies from one another and to invest money and other factors in business, we require a reasonably accurate forecast of demand. Hence, the forecaster must select that method that best suits the requirement. There is no particular method that enables organizations to anticipate risks and uncertainties in the future.

The methods of demand forecasting can be classified as:

  • QUALITATIVE/ SURVEY METHODS:
    1. CONSUMER SURVEY
    2. OPINION MODEL
    3. DELPHI MODEL
    4. NOMINAL GROUP TECHNIQUE
  • QUANTITATIVE/ STATISTICAL METHODS:
    1. TREND PROJECTION
    2. BAROMETRIC METHOD
    3. ECONOMETRIC METHOD
    4. OTHER STATISTICAL METHODS

These methods can be explained as under:

QUALITATIVE METHODS OF DEMAND FORECASTING:

The survey method or qualitative method is one of the most commonly used methods for forecasting demand in the short term. In this, organizations conduct surveys to determine demand directly from consumers. Some of the survey methods are :

Consumer survey:

This method includes direct contact with consumers to carry out a survey of what they prefer and intends to buy. The survey method depends upon the type of product and buyers for which the survey has to be conducted.

For example, if a product is consumer durable, a sample survey can be carried out about what they are planning to buy and intending to buy. If a product is sold to large industrial buyers, the survey would include interviewing them.   

 The consumer surveys can be through direct contact or questionnaire through the mail. These surveys build a relationship between

i)Price and demand

ii)Demand and income of consumers

Advertisement-X

iii)Demand and expenditure on the advertisement.

This method is useful when—– bulk sales made to industrial purchasers and only a few of them have to be contacted. This method is not useful for households as interviewing them is difficult as well as expensive.

Opinion model:

In this method, sales representatives predict the estimated future sales individually in their respective areas. The individual estimates are then aggregated to determine the total estimated future sales. The principle underlying this method is that salesmen are closest to the consumers and they have the intimate feel of the market. They are more likely to understand the reason behind the changes in consumers’ needs and demands. Thus, they are most suitable for assessing the consumer’s reaction to the company’s products. Therefore, a company having good sales personnel can utilize their experiences to predict the demands. Hence, this method is also called the Salesforce opinion or Grassroots approach method.

Although this method is simple, direct, first hand, easy and acceptable, it includes the following drawbacks——-

i) Salespersons may not prepare demand estimation with seriousness and care

ii) The salesperson may not have the required knowledge and experience

iii) Each salesperson has knowledge about a small portion of the market, thus, predicting total demand on this may be risky.

Delphi method:

This technique was developed at RAND Corporation in the 1950s. It is a group process where experts in the field of marketing research make demand forecasting. The experts are interrogated through a sequence of questionnaires in which responses to the first questionnaire are used to prepare the second questionnaire. The information available to some experts is shared with all experts for forecasting. This method is used for long term forecasting to estimate potential sales for new products. This method presumes two conditions:-

1)The panelists must be rich in expertise, knowledge, and experience.

2)The conductors are objective in their job. This method saves time and other resources.

Advertisement-X

Nominal Group Technique of Demand Forecasting:

It was originally developed by Delbecq and Vandeven. It is a further modification of the Delphi method. In this technique, a panel of 3-4 groups of up to 10 experts is created and allowed to interact, discuss, and rank all suggestions.

In the first phase of this technique, experts are asked to speak to each other while sitting together. The experts are asked to write down a list of ideas regarding the questions needing forecast. After writing down the ideas, the administrator asks each expert to share the best idea and show that on the flip chart. The ideas from all the experts are shown on the flip chart and about 15-20 ideas have emerged. In this phase, no discussion takes place whereas only ideas are examined.  

In the next phase, the experts discuss their respective ideas, and similar ideas are combined to minimize the number of ideas. After the discussion, the experts are asked to rank the ideas according to their perception of priority.

QUANTITATIVE/STATISTICAL METHODS OF DEMAND FORECASTING:

The statistical methods are used when forecasting is to be done for a longer period of time. These methods utilize the time series and cross-sectional data to estimate demand. these methods are considered more superior as compared to other methods due to the following reasons: 

1)The estimates are real. 

 2)There is a minimum subjectivity in these methods.

 3) The cost involved is minimal.

Methods are scientific and based on the relationship between the dependent and independent variables. some of the quantitative methods are :

1.Trend Projection:

Trend projection is the classical method of forecasting in business. In this method, the sales forecasts made through analysis of historical data taken from previous year’s books of accounts. This technique assumes that whatever past year’s demand pattern will be continued in the future too. The historical data are arranged chronologically yield what is referred to as ‘time-series’.

Time series data are composed of:

Advertisement-X

A) Secular Trend (T):

It refers to the long-run changes that occur as a result of a general tendency.

B) Seasonal Variation(S):

It refers to changes in short-run weather patterns or social habits.

C) Cyclical Element(C):

It refers to the changes that occur in the industry during boom and depression.

D) Random Variation (I):

It refers to the factors which are generally able such as wars, strikes, flood, and so on. This is the most popular method as it is simple and inexpensive.

 The trend projection includes more methods:

Graphical Method:

In this, the forecasting is done with the help of graphs. The sales belonging to previous years are plotted on a graph and a line is drawn on plotted points to know the trend in past years. This method is very simple and less expensive but data may be biased by the forecaster.   

Least Square Method:

In this, a trend line can be fitted to the time series data with the help of the least square regression. There are two types of trends are taken into account in this method, which are —

Linear Trend:

It implies the trend in which sales show a rising trend. In this, a straight line trend equation is fitted: S = A+Bt

where S = annual sales,

t = time (in years)

Advertisement-X

A and B are constant where B gives the measure of the annual increase in sales.  

Exponential Trend:

It implies the trend in which sales increase over the past years at an increasing or constant rate. In this, the trend equation be used as Y =aTb

Where Y = Annual sales,   

T= time in years 

a and b are constant.

 This method is very easy and inexpensive to use.

Barometric Technique:

The barometer is an instrument which measures changes. This method was introduced by Harvard Economic Service in 1920 and further revised by the National Bureau of Economic Research (NBER) in the 1930s. In this method, barometric techniques are used which are based on the idea that certain events of the present can be used to predict the pattern of changes in the future. this can be accomplished by the economic and statistical indicators such as savings, investment, and income which serve as a barometer of economic change.

Generally, forecasters correlate a firm’s sales with three series:  

i)The Leading Series:

The leading series includes those factors which move up or down before the recession or recovery phase of the business cycle starts. For example, the data relating to working women would act as a leading indicator for the demand of working women hostels. The most common examples for leading indicators are net business investment index, a new order for durable goods, change in the value of inventories, corporate profits after tax, etc. though these indicators provide a way to understand future demand, their major drawback is that they may not be always precise.  

ii) Coincident or Concurrent Series:

The coincidental series include indicators that move up or down simultaneously with the general level of economic activities. The most common examples for coincidental series are the rate of unemployment, sales by manufacturing, retail and trading sectors, gross national product at all prices.      

Advertisement-X

  iii) Lagging Series:

The lagging series comprises of the indicators which take place after happening. These indicators are very much important to understand how the economy would shape up in the future as these follow the economic cycle.  Inflation and data relating to unemployment levels are top indicators that help in analyzing the performance in the economy.

3. Econometric method:

This method is a tool that reveals relationships among economic variables to forecast future developments. It combines statistical tools with economic theories for forecasting. This method is regarded as more reliable than others. This model can either be a single equation regression model or may consist of a system of simultaneous equations. In most commodities, the single equation regression model is used but in the case where economic variables are so interdependent or interrelated to each other that unless one is defined, the other can’t be determined, and then a system of simultaneous equations is used to forecast. The economic method comprised of two basic methods:  

 a) Regression method:

It is the most commonly used method to forecast the demand for a product. This method combines the economic theory with statistical tools of estimation. The economic theory is used to specify the demand determinants and relation between product demand and its determinants. While the statistical techniques are used to estimate the value of parameters in the projected equation. Under this method, the first thing is to determine the demand function. While specifying the demand function, it is very important to understand that whether the demand depends by or large, on a single independent variable or multiple variables. If demand depends upon a single independent variable, such demand function is called a single variable demand function and a simple regression equation is used for forecasting. And if demand depends on multiple variables, such demand function is known as multiple variable functions and a multivariable equation is used for estimating the demand for a product.

 For instance, if it is found that in a city, the demand for necessity goods depends largely on the population of a city, then it would be regarded as a single variable demand function. On the other hand, if it is found that the demand for products like fruits, vegetables etc. depends on a number of variables such as the price of the product itself, prices of substitutes, the income of consumers, population, etc., it would be regarded as multi-variable demand function. 

b) Simultaneous equation Model:

In this model, demand forecasting involves the estimation of several simultaneous equations. These equations are generally the behavioral equations, market-clearing equations, and mathematical identities. This technique is based on the assumption that independent variables cause variation independent variable but not vice versa. In other words, the independent variables are no way affected by dependent variables. On the contrary, this model enables the forecaster to study the simultaneous interaction among dependent and independent variables. Thus, it is considered as a systematic and complete approach for forecasting as it employs several mathematical and statistical tools for estimation.

 4.Other Statistical Methods:

Apart from statistical tools, there are other methods for demand forecasting. These methods are very specific and used for particular data sets. These methods cannot be used for all types of research. These methods are :

a) Index Number:

It refers to the measures which are used to study the fluctuations in a variable or set of variables with respect to time. These are the most commonly used measures in economics and financial research to study factors such as price and quantity of products. The index number can be classified as:

i) Simple Index Number:

It refers to the number that measures the relative change in a single variable with respect to the base year.  

ii) Composite Index Number:

it refers to the number that measures a relative change in a set of variables with respect to the base year.

iii) Price Index Number:  

It refers to the number that measures the relative change in the price of a product in different time periods.  

Advertisement-X

iv) Quantity Index Number:

It refers to the number that measures the change in the physical quantity of goods produced, consumed, or sold in different time periods.

b) Time Series Analysis:

It refers to the analysis of a series of observations over a period of equally spaced time intervals. It is applicable in various fields such as the public sector, economics, and research. It comprises of:

i) Secular Trend:

It refers to a trend that is denoted by T and prevalent over a period of time. The secular trend for a data series can be upward or downward depending upon the trend. The upward trend shows an increase in the variables whereas the downward trend shows declining phases.

ii) Short Time Oscillation:

It refers to the trend that remains for a shorter period of time. It can be classified as:

 Seasonal Trend:

It refers to the trend that occurs year after year for a particular period. The reasons for such trends can be weather conditions, festivals, or some other customs. For example, an increase in the demand for sweets near Diwali or other festivals.

 Cyclical Trend:

It refers to the trend that lasts more than for a year. These trends are neither continuous nor seasonal in nature. For example, the business cycle.   

Irregular Trend:

It refers to the trend that is short and unpredictable. For example, volcano eruptions, floods, and earthquakes.         

c) Decision Tree Analysis:

It refers to the model that is used to make a decision in an organization. In this analysis, a tree-shaped structure is drawn to find out the best solution for a problem. In this, firstly, we have to find out different options that we can apply to solve a particular problem. Then, we will find out the outcome of each option. The flow of the decision tree should be from left to right. In the flow chart, the options are connected with a square nod, and outcomes are shown with a circle nod.

In short, there are several methods to determine future demand but which method or technique is used; it is an extremely difficult process for a business to carry out. Even with the wealth of data and a panel of full experts, trying to predict the future is not an easy task. Thus, while selecting methods for forecasting, it becomes necessary to consider all the factors whether internal or external.

Thanks for reading the topic

Advertisement-X

Please comment on your feedback in the comment box whatever you want. If you have any questions please ask us by commenting.

References:

Introductory Microeconomics – Class 11 – CBSE (2020-21) 

Advertisement

error: Content is protected !!