Stores the output of Bayesian Gaussian graphical model selection and
averaging, as produced by function modelSelectionGGM
.
The class extends a list, so all usual methods for lists also work for
msfit_ggm
objects, e.g. accessing elements, retrieving names etc.
Methods are provided to obtain parameter estimates, posterior intervals (Bayesian model averaging), and posterior probabilities of parameters being non-zero
Objects are created by a call to modelSelectionGGM
.
The class extends a list with elements:
Sparse matrix (dgCMatrix
) with posterior
samples for the Gaussian precision (inverse covariance)
parameters. Each row is a posterior sample. Within each row, only
the upper-diagonal of the precision matrix is stored in a flat
manner. The row and column indexes are stored in indexes
For each column in postSample, it indicates the row and column of the precision matrix
Number of variables
Priors specified when calling modelSelection
Obtain BMA posterior means, intervals and posterior probability of non-zeroes
Shows estimated posterior inclusion probability for each parameter vs. number of MCMC iterations. Only up to the first 5000 parameters are shown
signature(object = "msfit_ggm")
:
Displays general information about the object.
David Rossell
modelSelectionGGM