Demand Forecasting

OBJECTIVES

1. To understand the forecasting process when trend exists in demand.

2. To study a case involving downward trend of demand.

3. To learn the technique of trend projection.

4. To learn the technique of standard error of the estimate.

LEARNING FROM THE CLASS

The session gave us a brief description regarding the context of demand forecasting with explanation and

practical examples. It taught us about the concept of demand forecasting and its application in the

companies. Learned trend projection method, correlation and standard error of the estimate.

SUMMARY OF THE TOPICS

Demand forecasting means predicting future events or demand used in planning process.

We begin with different types of demand patterns.we examine forecasting method in three basic categories:-

A. Judgement

B. Casual

C. Time-series method

In thjs session we learnt following 6 topics-

 Trend projection

 Least square method

 Associative forecasting

 Correlation

 Coefficient of determination

 Multiple regression analysis

 Standard error of the estimate

We will see all these one by one in next page…...

 TREND PROJECTION

 Trend projection method is the most classical method of business forecasting, which is

concerned with movement of variables through time.

 This method requires a long time series data.

 Linear trends can be determined using the least square technique:

 LEAST SQUARE METHOD

 Least square method minimises the sum of the squared errors(deviation).

y=a+bx

where,

y=computed value of the variable to be predicted

a=y-axis intercept

b=slope of trend line

x=independent variable

 The deviations can be negative or positive.



 ASSOCIATIVE FORECASTING

 Used when changes in 1 or more independent variable can be used to predict the changes in the

dependent variable.

 CORRELATION

 Checks the bond between the variables.

 Denoted by r.

 Values range from -1 to +1.

 COEFFICIENT OF DETERMINATION

 Measures the percent of change is y predicted by change in x.

 Denoted by r 2 .



...