A forecast is a prediction or estimation of a future situation, under given conditions.
Forecasts can broadly be classified into categories:
(i) Passive forecasts
(ii) Active forecasts
(i) Passive forecasts: where prediction about future is based on the assumption that the firm does not change the course of its action.
(ii) Active forecasts: where forecasting is done under the condition of likely future changes in the actions by the firm
Generally, business firms are interested in both passive and active forecasts.
DETERMINING SCOPE OF A FORECASTING EXERCISE
Period of forecast
(i) Short-run forecasting: seasonal factors are the ingredients of short-run forecasts. Ex. Textiles and Apparel Industry, Electricity
(ii) Medium-term forecasting: Policies are the factors affecting this. Ex present State Government policy of banning the Cola drinks at the government schools and colleges
(iii) Long-term forecasting: various variables having inter-relationship of economic, psychological and sociological factors determining consumer behaviour. Ex diversification
Levels of forecast
(i) Macro-economic forecasting: Censes survey. Young India survey by CNN
(ii) Industry demand forecasting: Tata’s Westside had conducted a survey in 1997 about retail market before starting the Westside retail outlet on Dec 15 th 1998.
(iii) Firm demand forecasting: Godrej
(iv) Product-line forecasting: helps the firm to decide which of the product or products should have priority in the allocation of firm’s limited resources. P&G whether to produce more of hair shampoo( either head and shoulders or Pantene) or to produce Tide or Camay soap. Premium End segment
Qualities of Good Forecasting
2) Economy of time
3) Economy of money
METHODS OF DEMAND FORECASTING Opinion Polling Methods Statistical Methods Consumers’ Survey Method Sales Force Opinion Method Experts’ Opinion Method Mechanical Extrapolation (Trend Projection Method) Barometric Techniques Regression Method Complete Enumeration Survey Sample Survey End-use Leading, Lagging and Coincident indicators Diffusion indices Exponential Smoothing Fitting Trend Line by Observation Clinic approach Market Experiment Test Market Moving Averages
Complete Enumeration Survey:
DF = (ID 1 + ID 2 + ID 3 + ….ID n )
Where DF = demand forecast for all households, ID = intended demand of household. Ex. Census survey
Advantages: first-hand unbiased information
Disadvantages: costly, unwillingness of consumers to answer, uncertainty, time consuming
Simple Random Sampling
DF = (ID 1 + ID 2 + ID 3 + ….ID n ) N/n where N is population and n is sample picked up.
Advantages: takes less time and money
Disadvantage: unwillingness of consumers to answer, uncertainty, Biased answers
Test Market : This is actual experiment where shops are open in different localities and then consumers’ reactions are observed and recorded. Ex. Little heart biscuits (Britannia)
Clinic Method : or Simulation method or laboratory experiment, involves providing token money to a set of consumers and asking them to shop around in a simulated market.
End Use Method
Steps: identify the use of the intermediate product in the final product
Identify the demand in the national and international markets.
Then project the sale of the product under consideration.
Sales Force Opinion Method
When any Oil Engines manufacturing companies want to produce trolley, gathering information from the sales force is Sales force opinion method or collective opinion.
The men who are closest to the market are questioned and their responses aggregated.
Advantages: cheap and easy
Disadvantages: Congenital optimism and congenital pessimism, near-future forecasting is only possible, various socio-economic factors are not considered.
Experts’ Opinion Survey Method
Simple Method : Researchers identifies the ‘experts’ on the commodity whose demand forecast is being attempted and probes with them on the likely demand for the product in the forecast period.
Graphical Method : under this method, a graph of historical data on the variable under forecasting is drawn, it is then extrapolated visually upto the forecast period.
The level of sales in May, June, July, and August were 84, 92, 83, and 89, respectively. What is the four-period moving average forecast of sales in September?
Answer: [(84 + 92 + 83 + 89)/4 = 348/4 = 87]
S = a (this year’s sales) + (1- a) this year’s forecast
a is smoothing range which is from 0.0 to 1.0
Forecasted sales = 350 units, Actual sales = 320 units
If Smoothing constant is .3
.3 (320) + .7 (350) = 341 units of product
The forecast level of sales for the month of October was 140 units. Actual sales in October turned out to be 130 units. Use an exponential smoothing coefficient of 0.60 to forecast sales for November.
Answer: [(130)(0.60) + (140)(0.40) = 134]
When any economic variable is under forecast, the related leading indicators forecast is taken. This is based on the idea that future can be predicted from certain events occurring in the present.
Ex. Birth rate of children is the leading series for demand of seats in schools.
There three types of indicators
Leading indicators, coincident and lagging indicators.
Leading indicators: these are the variables whose movement precedes the movement of some other related variable. In the Estimation of GDP the leading indicator is aggregate investment in all the three sectors.
The world’s interest rate is the leading indicator for the credit expansion of contraction in India
To overcome some of problems of leading indicator method, the below two methods are developed.
Identify the leading indicators
If there are say 10 proper leading indicators for forecasting, by plotting it was found that 7 indices show a rise,
(7/10 x 100) = 70%. This means that when the index exceeds 50%, all the 7 indices are rising and so the variable under forecast will also have a upward trend.
The objective of the forecast: to estimate the export of good X
Find the various variables which will influence the export of good X.
Example National income (a), domestic prices of X (b), International prices of X (d) and whether (d).
Regression analysis was used to estimate the following linear trend equation:
St = 10.5 + 0.25 t
Use this equation to forecast the value of the dependent variable in time period 10.