rmixGibbs(bayesm) | R Documentation |
rmixGibbs
makes one draw using the Gibbs Sampler for a mixture of multivariate normals.
rmixGibbs(y, Bbar, A, nu, V, a, p, z, comps)
y |
data array - rows are obs |
Bbar |
prior mean for mean vector of each norm comp |
A |
prior precision parameter |
nu |
prior d.f. parm |
V |
prior location matrix for covariance priro |
a |
Dirichlet prior parms |
p |
prior prob of each mixture component |
z |
component identities for each observation – "indicators" |
comps |
list of components for the normal mixture |
rmixGibbs
is not designed to be called directly. Instead, use rnmixGibbs
wrapper function.
a list containing:
p |
draw mixture probilities |
z |
draw of indicators of each component |
comps |
new draw of normal component parameters |
This routine is a utility routine that does not check the input arguments for proper dimensions and type.
Rob McCulloch and Peter Rossi, Graduate School of Business, University of Chicago, Peter.Rossi@ChicagoGsb.edu.
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