Bayesian I-optimal designs for choice experiments with mixtures

Discrete choice experiments are frequently used to quantify consumer preferences by having respondents choose between different alternatives. Choice involving mixtures of ingredients have been largely overlooked in the literature, even though many products and services can be described as ingredients. As a consequence, little research has done on optimal design mixtures. The only existing focused D-optimal designs, which means that an estimation-based approach was adopted. However, with mixtures, it is crucial obtain models yield precise predictions for any combination ingredient proportions. This because goal mixture generally find optimizes respondents’ utility. result, I-optimality criterion more suitable designing than D-optimality focuses getting estimated statistical model. In this paper, we study Bayesian I-optimal compare them their counterparts, show former designs perform substantially better latter terms variance predicted