Jeffrey′s Noninformative Prior in Bayesian Conjoint Analysis

  • Oh, Man-Suk (Ewha Womans University) ;
  • Kim, Yura (IT Service Team/Information Analysis Part, LG Capital)
  • Published : 2000.06.01

Abstract

Conjoint analysis is a widely-used statistical technique for measuring relative importance that individual place on the product's attributes. Despsite its practical importance, the complexity of conjoint model makes it difficult to analyze. In this paper, w consider a Bayesian approach using Jeffrey's noninformative prior. We derive Jeffrey's prior and give a sufficient condition under which the posterior derived from the Jeffrey's prior is paper.

References

  1. Marketing Science v.15 Hierarchical Bayes Conjoint Analysis;Recovery of Partworth Heterogenetiy from Reduced Experimental Desings Lenk, Peter, J.;DeSarbo, Wayne S.;Green, Paul E.;Young, Martin R.
  2. Matrix Differential Calculus with Applications in Statistics and Econometrics Magnus, J.R.;Neudecker, H.
  3. Aspects of Multivariate Statistical Theory Muirhead, R.J.
  4. Canadian Journal of Statistics v.16 Patterned Matrix Derivatives Tracy, D.S.;Jinadasa, K.G.
  5. Linear Algebra Application v.67 On some pattern-reduction matrices which appear in statistics Wiens, D.P.
  6. Journal of Multivariate Analysis v.55 Bayesian Analysis for Random Coefficient Regression Models Using Noninformative Priors Yang, Ruoyong;Chen, Ming-Hui