The Application of Discriminant Analysis to the Prediction of Sales Forecast Uncertainty in New Product Situations

Abstract

Managers' estimates of sales forecast uncertainty for a new product are critical to an effective risk analysis. This research utilizes discriminant analysis to analyze sales forecast errors in past new-product ventures. Using the past ventures, a linear discriminant function is established from relevant market variables by which future ventures may be allocated to one of two populations depending on the probabilities that their sales forecast uncertainty will be low or high.