Bayesian analysis of a binary choice multidimensional scaling model with correlated errors using the Gibbs sampling method
University of Michigan, 1995 - Mathematics - 234 pages
What people are saying - Write a review
We haven't found any reviews in the usual places.
IH THE SPATIAL REPRESENTATION OF PICK ANYJ DATA
CONSUMER SOFT DRINK
AMEAN analyze approach approximate asymptotic Bayes factor Bayesian analysis brand coordinates brand location CALCULATE CHOC choice behavior choice model choice probabilities choice uncertainty CHOLESKY ROOT computing conditional distributions CONTINUE ENDIF convergence correlated errors correlation matrix covariance matrix data augmentation DeSarbo and Cho Diet 7Up Diet Coke Diet Pepsi dimension dimensional ergodic error terms fx(x Gelfand and Smith Gibbs Draws Gibbs estimated Gibbs sampler Gibbs sampling algorithm IMPLICIT REAL*8 A-H independence assumption individual latent class latent utility likelihood function log Bayes factor marginal distribution Markov chain matching coefficient McCulloch and Rossi MDIM Means and Standard model with correlated Monte Carlo Mountain Dew multidimensional multivariate normal distribution NDIM NTDIM Pepsi Light phi coefficient posterior distribution posterior mean proposed model quantiles random variables sampling-based Sequence of Gibbs simulated soft drink soft drink data Sprite standard deviation Statistical subject vectors SUBROUTINE threshold values truncated multivariate normal WISHART DISTRIBUTION WRITE(NOU