Previous conjoint choice design construction procedures have produced a single design that is administered to all subjects. This paper proposes to construct a limited set of different designs. The designs are constructed in a Bayesian fashion, taking into account prior uncertainty about the parameter values. A computational procedure is developed that enables fast and easy implementation in practice. Even though the number of such different designs in the optimal set is small, it is demonstrated through a Monte Carlo study that substantial gains in efficiency are achieved over aggregate designs.

consumer preferences, discrete choice, estimator efficiency, experiments, multinomial logit
Econometric and Statistical Methods: General (jel C10), Statistical Decision Theory; Operations Research (jel C44), Business Administration and Business Economics; Marketing; Accounting (jel M), Marketing (jel M31)
hdl.handle.net/1765/320
ERIM Report Series Research in Management
Erasmus Research Institute of Management

Sándor, Z, & Wedel, M. (2003). Differentiated Bayesian Conjoint Choice Designs (No. ERS-2003-016-MKT). ERIM Report Series Research in Management. Retrieved from http://hdl.handle.net/1765/320