Minimum aberration designs for discrete choice experiments
A discrete choice experiment (DCE) is a survey method that gives insight into individual preferences for particular attributes. Traditionally, methods for constructing DCEs focus on identifying the individual effect of each attribute (a main effect). However, an interaction effect between two attributes (a two-factor interaction) better represents real-life trade-offs, and provides us a better understanding of subjects’ competing preferences. In practice it is often unknown which two-factor interactions are significant. To address the uncertainty, we propose the use of minimum aberration blocked designs to construct DCEs. Such designs maximize the number of models with estimable two-factor interactions in a DCE with two-level attributes. We further extend the minimum aberration criteria to DCEs with mixed-level attributes and develop some general theoretical results.
KeywordsAliasing blocked design estimation capacity fractional factorial design multinomial logit model orthogonal array
AMS Subject Classification62K15 62J15 62K10
Unable to display preview. Download preview PDF.
- Bliemer, M., and J. Rose. 2011. Experimental design influences on stated choice output: An empirical study in air travel choice. Transportation Research Part A: Policy and Practice 45 (1):63–79.Google Scholar
- Bunch, D. S., J. J. Louviere, and D. Anderson. 1996. A comparison of experimental design strategies for choice-based conjoint analysis with generic-attribute multinomial logit models. Working paper, Graduate School of Management, University of California, Davis, California.Google Scholar
- Gerard, K., M. Shanahan, and J. Louviere. 2008. Using discrete choice modeling to investigate breast screening participation. In Using discrete choice experiments to value health and health care, ed. M. Ryan, K. Gerard, and M. Amaya-Amaya, Dordrecht, The Netherlands: 117–137.Google Scholar
- Groemping, U., B. Amarov, and H. Xu. 2015. DoE.base: R package version 0.27-1. https://doi.org/CRAN.R-project.org/package=DoE.base. August 21.
- Johnson, F. R., E. Lancsar, D. Marshall, V. Kilambi, A. Muhlbacher, D. A. Regier, B. W. Bresnahan, B. Kanninen, and J. F. P. Bridges. 2013. Constructing experimental designs for discrete-choice experiments: Report of the ISPOR Conjoint Analysis Experimental Design Good Research Practices Task Force. Value in Health 16:3–13.CrossRefGoogle Scholar
- Xu, H. 2015. Nonregular factorial and supersaturated designs. In Handbook of design and analysis of experiments, ed. A. Dean, M. Morris, J. Stufken, and D. Bingham, 339–70. Boca Raton, FL: Chapman and Hall/CRC.Google Scholar
- Xu, H., and R. W. Mee. 2010. Minimum aberration blocking schemes for 128-run designs. Journal of Statistical Planning and Inference 140:3213–29. Xu, H., and C. F. J. Wu. 2001. Generalized minimum aberration for asymmetrical fractional factorial designs. The Annals of Statistics 29 (4):1066–77.MathSciNetCrossRefGoogle Scholar