R: Compute Posterior Expectation of Normal Mixture Model Moments

momMix {bayesm}

R Documentation

Compute Posterior Expectation of Normal Mixture Model Moments

momMix averages the moments of a normal mixture model over MCMC draws.

momMix(probdraw, compdraw)

`probdraw`	`R x ncomp` list of draws of mixture probs
`compdraw`	list of length `R` of draws of mixture component moments

`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 `i`th draw.
`compdraw[[i]][[j]][[1]]`	is the mean parameter vector for the `j`th component, `i`th MCMC draw.
`compdraw[[i]][[j]][[2]]`	is the UL decomposition of `\Sigma^{-1}` for the `j`th component, `i`th MCMC draw

A list containing:

`mu`	posterior expectation of mean
`sigma`	posterior expectation of covariance matrix
`sd`	posterior expectation of vector of standard deviations
`corr`	posterior expectation of correlation matrix

This routine is a utility routine that does not check the input arguments for proper dimensions and type.

Peter Rossi, Anderson School, UCLA, perossichi@gmail.com.

For further discussion, see Chapter 5, Bayesian Statistics and Marketing by Rossi, Allenby, and McCulloch.