Multimodal preferene heterogeneity in choice-based conjoint analysis: a simulation study

Abstract The most commonly used variant of conjoint analysis is choice-based (CBC). Here, hierarchical Bayesian (HB) multinomial logit (MNL) models are widely for preference estimation at the individual respondent level. A new and very flexible approach to address multimodal skewed heterogeneity in context CBC Dirichlet Process Mixture (DPM) MNL model. number masses components do not have be predisposed like latent class (LC) model or mixture-of-normals (MoN) aim this Monte Carlo study evaluate performance choice (basic MNL, HB-MNL, MoN-MNL, LC-MNL DPM-MNL models) under varying data conditions (especially structures) using statistical criteria parameter recovery, goodness-of-fit predictive accuracy. core finding from that standard HB-MNL appears highly robust settings .