Although the panel consensus method provides forecasts that involve collective viewpoints of experts, however such forecasts can be biased as it can be affected by strong personality of one or more key individuals.
To alleviate such problems the method adopted by BCFF is similar to that of the Delphi Method. In the Delphi method, members of the panel of experts respond to a series of questions related to the forecasting problem. An independent party to draw out a consensus opinion out of it analyzes the responses.
Such an approach followed by BCFF to make forecasts reduces the steamroller or bandwagon problems of the panel discussion method as it reduces the risk of biased forecasts due to group pressures which otherwise could be derived out of regular group meetings.
a) For TUL, the forecasts of Christmas season sales should be based on the following equation:
Here is the forecast for Christmas season sales is last season sales and
S is the change in sales.
Based on the company estimates
S can be written as:
is the rate of change of prices and
is the rate of change of customer traffic.
Substituting Equation (2) in Equation (1),
The forecast of Christmas sales for TUL should be based on Equation (3).
b) For the given the values of variables in Equation (3) the forecast for Christmas season sales are as follows:
Hence, for the given data forecast for this season Christmas sales are 12,500 games.
A linear regression model of advertising-sales relationship for forecasting advertising levels at which threshold or saturation effects become prevalent will not be appropriate as there is an irregular relationship between the two and hence non-linear method of estimation is appropriate. Quadratic and log-linear models are often used for this purpose.