sampler_control {mcmcsae}

Set computational options for the sampling algorithms

Description

Set computational options for the sampling algorithms

Usage

sampler_control(
  add.outer.R = TRUE,
  recompute.e = TRUE,
  expanded.cMVN.sampler = FALSE,
  CG = NULL,
  block = TRUE,
  block.V = TRUE,
  auto.order.block = TRUE,
  chol.control = chol_control(),
  max.size.cps.template = 100,
  PG.approx = TRUE,
  PG.approx.m = -2L,
  CRT.approx.m = 20L
)

Arguments

add.outer.R

whether to add the outer product of a constraint matrix for a better conditioned linear system of equations, typically for coefficients sampled in a Gibbs-block. Default is TRUE. If NULL, a simple heuristic is used to decide whether to add the outer product of possibly a submatrix of the constraint matrix.

recompute.e

when FALSE, residuals or linear predictors are only computed at the start of the simulation. This may give a modest speedup but in some cases may be less accurate due to round-off error accumulation. Default is TRUE.

expanded.cMVN.sampler

whether an expanded linear system including dual variables is used for equality constrained multivariate normal sampling. If set to TRUE this may improve the performance of the blocked Gibbs sampler in case of a large number of equality constraints, typically identifiability constraints for GMRFs.

CG

use a conjugate gradient iterative algorithm instead of Cholesky updates for sampling the model's coefficients. This must be a list with possible components max.it, stop.criterion, verbose, preconditioner and scale. See the help for function CG_control, which can be used to specify these options. Conjugate gradient sampling is currently an experimental feature that can be used for blocked Gibbs sampling but with some limitations.

block

if TRUE, the default, all coefficients are sampled in a single block. Alternatively, a list of character vectors with names of model components whose coefficients should be sampled together in blocks.

block.V

if TRUE, the default, all coefficients of reg and gen components in a variance model formula are sampled in a single block. Alternatively, a list of character vectors with names of model components whose coefficients should be sampled together in blocks.

auto.order.block

whether Gibbs blocks should be ordered automatically in such a way that those with the most sparse design matrices come first. This way of ordering can make Cholesky updates more efficient.

chol.control

options for Cholesky decomposition, see chol_control.

max.size.cps.template

maximum allowed size in MB of the sparse matrix serving as a template for the sparse symmetric crossproduct X'QX of a dgCMatrix X, where Q is a diagonal matrix subject to change.

PG.approx

whether Polya-Gamma draws for logistic binomial models are approximated by a hybrid gamma convolution approach. If not, BayesLogit::rpg is used, which is exact for some values of the shape parameter.

PG.approx.m

if PG.approx=TRUE, the number of explicit gamma draws in the sum-of-gammas representation of the Polya-Gamma distribution. The remainder (infinite) convolution is approximated by a single moment-matching gamma draw. Special values are: -2L for a default choice depending on the value of the shape parameter balancing performance and accuracy, -1L for a moment-matching normal approximation, and 0L for a moment-matching gamma approximation.

CRT.approx.m

scalar integer specifying the degree of approximation to sampling from a Chinese Restaurant Table distribution. The approximation is based on Le Cam's theorem. Larger values yield a slower but more accurate sampler.

Value

A list with specified computational options used by various sampling functions.

References

D. Bates, M. Maechler, B. Bolker and S.C. Walker (2015). Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software 67(1), 1-48.

Y. Chen, T.A. Davis, W.W. Hager and S. Rajamanickam (2008). Algorithm 887: CHOLMOD, supernodal sparse Cholesky factorization and update/downdate. ACM Transactions on Mathematical Software 35(3), 1-14.