Generate approximate posterior samples from the sparse precision matrix Q, assuming multivariate normality.

Generate approximate posterior samples from the sparse precision matrix Q, assuming multivariate normality.

Usage

sample_Q(fit, nsamples = 4000, seed = NULL)

Arguments

fit

An object returned by sample_snuts

nsamples

The number of samples to draw, defaulting to 4000.

seed

An optional random seed

Value

A data.frame of approximate samples

Examples

fit <- readRDS(system.file('examples', 'fit.RDS', package='SparseNUTS'))
x <- sample_Q(fit)