Demand Forecasting

The opinion poll methods:

The opinion poll methods aim at collecting opinion of those who are supposed to possess knowledge of the market e.g. sales representative, professional marketing experts and consultants. The opinion poll method include

1.     Expert opinion method: - Firms having a good network of sales representative can put them to work of assessing the demand for the product in the areas that they represent. Sales representative, beings in close touch with the consumers are supposed to know the future purchase plans of their customer, their reaction to the market changes, their response to the introduction of new products and the demand for competing products. They are, therefore, in a position to provide an estimate of likely demand for their firm’s product in the area. The estimates of demand thus obtained from different regions are added up to get the overall probable demand for a product.

2.     Delphi Method: - Delphi method is used to consolidate the divergent expert opinions and arrived at a compromise estimate of future demand.

Under Delphi method the expert are provided information on estimates of forecast of other experts along with the underlying assumptions. The experts may revise their own estimates in the light of forecast made by other experts. The consensus of experts about the forecasts constitutes the final forecast.

Although this method is simple and inexpensive, it has its own limitations. First estimates provided by sales representations and professional experts are reliable only to extend depending upon their skill to analysis the market and their experience. Second, demand estimates way involve the subjective judgement of the which may lead to over or under estimation, finally, the assessment of market demand is usually based on inadequate information’s, such as changes in GNP, available of credit, future prospects of the industry etc, fall outside their purview.

3.     Market studies and Experiments:- It is a method of collecting necessary information regarding demand is to carry out market studies and experiments on consumer’s behavior under actual through controlled market conditions. This method is known in common parlance market conditions. This methods is known in common parlance as market experiment method under this method, firms first select some areas of the representative markets – three or four cities having similar features viz. Population, income levels, cultural and social background, occupational distribution, choices and preferences of consumers. Then, they carry out market experiments by changing prices, advt. Expenditure and other controllable variable in the demand function under the assumption that other thing remains same. The controlled variable may by changed over time either simultaneously in all the markets or in all the markets or in the selected markets. After such changes are introduced in the market, the consequent changes in the demand over a period of time (a week, a fortnight or month) are recorded. On the basis of data collected elasticity coefficient are computed. These coefficients are then used along with the variables of the demand function to assess the demand for product

The market experiments methods have certain serious limitations. First, this method is very expensive and hence cannot be afforded by small forms. Second, being a costly affair, experiments are usually carried out on a scale too small to permit generalization with a high degree of reliability.

Third experimental methods are based on short – term and controlled conditions that may exist in an uncontrolled market. Hence, the results may not be applicable to the uncontrolled long-term conditions of the market.

Statistical Methods:

Trends Projection Method

Trend projection method is a classical method of business forecasting. This method is essentially concerned with the study of movement of variable through time. The use of this method requires a long and reliable time series data. The trend projection method is used under the assumption that the factors responsible for the past trends in variables to be projected (e.g. sales and demand) will continue to play their part in future in the same manner and to the same extend as they did in the past in determining the magnitude and direction of the variable.

There are three (3) techniques of trend projection based on time – series data.

1.     Graphical Method: - under this method, annual sales data is plotted on a graph paper and a line is drawn through the plotted points. Then a free hand line is so drawn that the total distance between the line and the point is minimum. Although this method is very simple and least expensive, the projections made through this method are not very reliable. The reason is that the extension of the trend line involves subjectivity and personal bias of the analysis.

2.     Fitting Trend Equation: Least square method: - Fitting trend equation is a formal technique of projecting the trend in demand. Under this method, a trend line (or curve) is fitted to the time – series data with the aid of statistical techniques. The form of the trend equation that can be fitted to the time series data is determined either by plotting the sales data or by trying different forms of trend equations for the best fit.

o    When plotted, a time series date may show various trends. The most common types of trend equation are

1) Liner and
2) exponential trends

• Linear Trend: - When a time series data reveals a rising trend in sales than a straight-line trend equation of the following form is fitted. (S = A + BT ; Where S = annual sales , T = Time (in year) , A & B are constant. The parameter b given the measure of annual increase in sales)

• Exponential trend:- When sales ( or any dependent variable) have increased over the past years at an increasing rate or at a constant percentage rate, than the appropriate trend equation to be used is an exponential trend equation of any of the following type ( Y = aebt , Or its semi – logarithmic form -> Log y = = log a + bt; This form of trend equation is used when growth rate is constant.)

3.     Moving Averages Method:    In statistics, a moving average, also called rolling average, rolling mean or running average, is a type of finite impulse response filter used to analyze a set of data points by creating a series of averages of different subsets of the full data set.

·         Given a series of numbers, and a fixed subset size, the moving average can be obtained. The average of the first subset of numbers is calculated. The fixed subset is moved forward to the new subset of numbers, and its average is calculated. The process is repeated over the entire data series. The plot line connecting all the (fixed) averages is the moving average. Thus, a moving average is not a single number, but it is a set of numbers, each of which is the average of the corresponding subset of a larger set of data points. A moving average may also use unequal weights for each data value in the subset to emphasize particular values in the subset.

·         A moving average is commonly used with time series data to smooth out short-term fluctuations and highlight longer-term trends or cycles. The threshold between short-term and long-term depends on the application, and the parameters of the moving average will be set accordingly. For example, it is often used in technical analysis of financial data, like stock prices, returns or trading volumes. It is also used in economics to examine gross domestic product, employment or other macroeconomic time series. Mathematically, a moving average is a type of convolution and so it is also similar to the low-pass filter used in signal processing. When used with non-time series data, a moving average simply acts as a generic smoothing operation without any specific connection to time, although typically some kind of ordering is implied.

Limitation

The first limitations of this method arise out of the assumption that the past rate of change in the dependent variable will persist in the future too. Therefore, the forecast based on this method may be considered to be reliable only for the period during which this assumption holds.

Second, this method cannot be used for short-term estimates. Also it cannot be used where trend is cyclical with sharp turning points of trough and perks.

Barometric Forecasting Techniques:

Barometric techniques examine the relationships between causal or coincident events to predict future events. This approach is based on the logic that key current developments can serve as a barometer of the future. This approach assumes the key developments can be identified, measured and recorded as a statistical time series. The barometric or what is also called the leading indicators approach to forecasting is often traced to work done at the National Bureau of Economic Research from the 1920s through the 1940s.

A leading indicator predicts three to six months in the future another event. Examples of indicators include: payroll employment, personal income less transfer payments, an index of industrial production, stock prices, changes in business inventories, consumer expectations, building permits, new orders for goods and materials and retail sale

Regression analysis method:

The moving average method does not respond well to a time series that increases or decreases with time. Here we include a linear trend term in the model. The regression method approximates the model by constructing a linear equation that provides the least squares fit to the last m observations.

Simulated interaction:

Simulated interaction is a form of role playing for predicting decisions by people who are interacting with others. It is especially useful when the situation involves conflict. For example, one might wish to forecast how best to secure an exclusive distribution arrangement with a major supplier.

To use simulated interaction, an administrator prepares a description of the target situation, describes the main protagonists’ roles, and provides a list of possible decisions. Role players adopt a role and read about the situation. They then improvise realistic interactions with the other role players until they reach a decision; for example to sign a trial one-year exclusive distribution agreement. The role players’ decisions are used to make the forecast.

Using eight conflict situations, Green (2005) found that forecasts from simulated interactions were substantially more accurate than can be obtained from unaided judgement. Simulated interaction can also help to maintain secrecy. Information on simulated interaction is available from conflictforecasting.com.