Learn R Programming

growcurves (version 0.2.4.1)

mmCmvplusDpPost: Bayesian mixed effects model with a DP prior on by-subject effects and CAR prior on a multivariate set of multiple membership effects

Description

An internal function to dpgrowmm

Usage

mmCmvplusDpPost(y, X, Z, H, Wcase, Wsubject, Omega, omegaplus, groups, subjects, niter, nburn, nthin, strength.mm, corsess, shapealph, ratebeta, typemm)

Arguments

y
An N x 1 response (of subject-measure cases)
X
Fixed effects design matrix
Z
Random effects design matrix. Assumed grouped by subjects
H
Multivariate MM effects design matrix.
Wcase
An N x 1 multiple membership weight matrix to map supplemental random effects
Wsubject
An P.aff x S multiple membership weight matrix with rows equal to number of unique affected subjects
Omega
An S x S unnormalized adjacency matrix with entries equal to 1 where two effects communicate and 0, otherwise. Diagonal elements are zero
omegaplus
S x 1 vector of row sums of Omega
groups
S x 1 vector of group identifiers for each effect. Effects within each group communicate. Effects don't communicate across groups. Not used under "mmi" prior (though input is required).
subjects
An N x 1 set of subject identifiers
niter
The number of MCMC iterations
nburn
The number of MCMC burn-in iterations to discard
nthin
The step increment of MCMC samples to return
strength.mm
The shape and rate parameters for the $\Gamma$ prior on the CAR precision parameter, $\tau_{\gamma}$.
corsess
A single value to set the prior correlations among the multivariate q = ncol(H) orders for the MM effects. where $\tau_{\gamma}$ is replaced by the q x q, $\Lambda$.
shapealph
The shape parameter for the $\Gamma$ prior on the DP concentration parameter. The rate parameter is set of 1.
ratebeta
The rate parameter for the $\Gamma$ prior on the DP concentration parameter. Default value is 1.
typemm
An indicator the prior formulation specified for the multivariate MM effects term. Set typemm = 0 for "mmi" and typemm = 1 for "mmcar".

Value

res A list object containing MCMC runs for all model parameters.

See Also

dpgrow