Resource title

Optimization conjoint models for consumer heterogeneity

Resource image

image for OpenScout resource :: Optimization conjoint models for consumer heterogeneity

Resource description

A new optimization based method for capturing consumer heterogeneity for conjoint analysis is presented. The method is based on recently proposed individual consumer preference models using polyhedral optimization methods and statistical learning theory. The method is experimentally compared with Hierarchical Bayes (HB) using standard, widely used, simulation data. The experiments show that the proposed method matches and often outperforms HB in terms of accuracy. Moreover estimation is computationally very efficient, therefore the proposed method can be used for very large datasets. Finally, the method can also be used, like the existing individual preference models it is based on, for modelling better than HB interactions among product attributes.

Resource author

Resource publisher

Resource publish date

Resource language


Resource content type


Resource resource URL

Resource license

Copyright INSEAD. All rights reserved