A formal procedure for synthesizing the variety of predictive models usually available to the market research analyst is presented as a practical method. Two case studies indicate its feasibility. The method is Bayesian, involving the assignment of subjective probabilities to the various predictive models and the revision of these prior model probabilities according to a Dirichlet process if a set of performance data is available.
Get full access to this article
View all access options for this article.
References
1.
BatesJ. M. and GrangerC. W. J. (1969), “The Combination of Forecasts,” Operational Research Quarterly, 20, 451–68.
2.
BunnD. W. (1975), “A Bayesian Approach to the Linear Combination of Forecasts,” Operational Research Quarterly, 26, 325–9.
3.
BunnD. W. (1977), “A Comparative Evaluation of the Outperformance and Minimum Variance Bayesian Procedures for the Linear Synthesis of Forecasts,” Operational Research Quarterly, 28, 632–62.
4.
ChambersJ., MullickS., and SmithD. (1971), “How to Choose the Best Forecast,” Harvard Business Review, 65:4, 45–74.
5.
GeurtsM. D. and IbrahimI. B. (1975), “Comparing the Box-Jenkins Approach with the Exponentially Smoothed Forecasting Model Application to Hawaii Tourists,” Journal of Marketing, 12, 182–8.
6.
LindleyD. V. (1971), Making Decisions. New York: John Wiley & Sons, Inc.
7.
NewboldP. and GrangerC. W. J. (1974), “Experience with Forecasting Univariate Time Series and the Combination of Forecasts,” Journal of the Royal Statistical Society A, 137, 131–65.
8.
ParkerG., and SeguraE. (1971), “How to Get a Better Forecast,” Harvard Business Review, 12, 99–109.
9.
TheilH. (1968), Applied Economic Forecasts. Amsterdam: North Holland.
