Dick R. Wittink is the General George Rogers Clark Professor of Marketing and Management at Yale University. He received his Ph.D. from Purdue University, where he was briefly exposed to the early conjoint literature. He joined the faculty at Stanford University in 1975 and started several research projects in conjoint analysis there (his first working paper on conjoint, a simulation study of alternative metric- and nonmetric estimation methods, was completed in 1976). He has been an irregular contributor to the conjoint literature ever since, and is probably best known for several articles on the commercial use of conjoint analysis and many papers on the number-of-attribute-levels-effect. A current paper focuses on how this artificial effect can be minimized, if not eliminated.
In forecasting his research centers around the idea that there are opportunities to improve the conduct of predictive accuracy tests. For a demonstration of structural superiority for a model with great complexity, the aggregation of predictions â€“ across individuals, across time periods â€“ is useful to minimize the impact of unreliability. When such aggregation is impossible, he favors the application of diagnostic predictive accuracy. This concept is based on the idea that the demonstration of superior predictive accuracy often requires explicit consideration of the conditions under which superiority is achievable.
He is an area editor of Marketing Science, an associate editor of the International Journal of Forecasting , and an editorial board member of the Journal of Marketing, the Journal of Marketing Research, and the International Journal of Research in Marketing.