momMix: Compute Posterior Expectation of Normal Mixture Model Moments
Description
momMix
averages the moments of a normal mixture model over MCMC draws.Usage
momMix(probdraw, compdraw)
Arguments
probdraw
R x ncomp list of draws of mixture probs
compdraw
list of length R of draws of mixture component moments
Value
- a list of the following items ...
- muPosterior Expectation of Mean
- sigmaPosterior Expecation of Covariance Matrix
- sdPosterior Expectation of Vector of Standard Deviations
- corrPosterior Expectation of Correlation Matrix
concept
- mcmc
- normal mixture
- posterior moments
Warning
This routine is a utility routine that does not check the
input arguments for proper dimensions and type.Details
R is the number of MCMC draws in argument list above.
ncomp is the number of mixture components fitted.
compdraw is a list of lists of lists with mixture components.
compdraw[[i]] is ith draw.
compdraw[[i]][[j]][[1]] is the mean parameter vector for the jth component, ith MCMC draw.
compdraw[[i]][[j]][[2]] is the UL decomposition of $Sigma^{-1}$ for the jth component, ith MCMC draw.References
For further discussion, see Bayesian Statistics and Marketing
by Allenby, McCulloch, and Rossi, Chapter 5.
http://gsbwww.uchicago.edu/fac/peter.rossi/research/bsm.html