An Empirical Comparison of the Predictive Validity of Self-Explicated, Huber-Hybrid, Traditional Conjoint, and Hybrid Conjoint Models

Abstract

The authors compare the predictive validity of different types of self-explicated, Huber-hybrid, traditional conjoint, and hybrid conjoint models. The results indicate that traditional and hybrid conjoint models outperform the self-explicated and Huber-hybrid models but are comparable in predictive validity terms. A sharp drop in validity measures from calibration to validation predictions is observed for all the models except those of the self-explicated category.