Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Rejoinder: the usefulness of Bayesian optimal designs for discrete choice experiments

Rejoinder: the usefulness of Bayesian optimal designs for discrete choice experiments Limitations of the case study In their case study, Kessels et al. focused on one specific scenario involving six two‐level attributes and eight choice sets of size 2. They assumed homogeneous respondents and, hence, used the conditional logit model for their computations. Obviously, it would be bad practice to take one case study and extend the inference to all possible design scenarios. Nevertheless, the results obtained by Kessels et al. are very suggestive of the factors that determine the conditions in which Bayesian optimal choice designs prove better than utility‐neutral designs. When the prior variance is relatively small and the distance of the mean of the prior distribution from the zero vector is relatively large, we expect Bayesian designs to perform much better than utility‐neutral designs in terms of the Bayesian D ‐optimality criterion. We should, however, also point out that the case study conducted by Kessels et al. complements a wide range of other studies comparing Bayesian optimal designs for choice experiments, on the one hand, and orthogonal designs or utility‐neutral designs, on the other hand. Ferrini and Scarpa , for instance, performed a case study to compare various design approaches for choice experiments and concluded that http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Stochastic Models in Business and Industry Wiley

Rejoinder: the usefulness of Bayesian optimal designs for discrete choice experiments

Loading next page...
 
/lp/wiley/rejoinder-the-usefulness-of-bayesian-optimal-designs-for-discrete-yUHJ6ySA0F

References (30)

Publisher
Wiley
Copyright
Copyright © 2011 John Wiley & Sons, Ltd.
ISSN
1524-1904
eISSN
1526-4025
DOI
10.1002/asmb.903
Publisher site
See Article on Publisher Site

Abstract

Limitations of the case study In their case study, Kessels et al. focused on one specific scenario involving six two‐level attributes and eight choice sets of size 2. They assumed homogeneous respondents and, hence, used the conditional logit model for their computations. Obviously, it would be bad practice to take one case study and extend the inference to all possible design scenarios. Nevertheless, the results obtained by Kessels et al. are very suggestive of the factors that determine the conditions in which Bayesian optimal choice designs prove better than utility‐neutral designs. When the prior variance is relatively small and the distance of the mean of the prior distribution from the zero vector is relatively large, we expect Bayesian designs to perform much better than utility‐neutral designs in terms of the Bayesian D ‐optimality criterion. We should, however, also point out that the case study conducted by Kessels et al. complements a wide range of other studies comparing Bayesian optimal designs for choice experiments, on the one hand, and orthogonal designs or utility‐neutral designs, on the other hand. Ferrini and Scarpa , for instance, performed a case study to compare various design approaches for choice experiments and concluded that

Journal

Applied Stochastic Models in Business and IndustryWiley

Published: May 1, 2011

There are no references for this article.