GibbsSampler: The Metropolized restricted Gibbs sampler

In Kaukol/IBGS: Variable selection in ultrahigh dimensions using Gibbs sampler

The Metropolized restricted Gibbs sampler

Description

The Metropolized restricted Gibbs sampler

Usage

GibbsSampler(
  y,
  x,
  n.vars = ncol(x),
  perm = TRUE,
  n.models = 10,
  tau = 0.9,
  len = 1000,
  k = 1,
  gamma = 0.5,
  info = c("AIC", "BIC", "AICc", "exBIC"),
  family = c("gaussian", "poisson", "binomial")
)

Arguments

y	the response variable
x	the predictors
n.vars	the number of maximal predictors included in the candidate model
perm	the permutation of Gibbs sampler, default TRUE
n.models	the number of top selected models
tau	the threshold to select the important predictors in second step, default is 0.9
len	the half number of generated samples, default is 1000
k	the tuning parameter, default is 1
gamma	the parameter for extended BIC, default is 0.5
info	the selected model selection criterion from AIC, AICc, BIC and exBIC
family	the type of model from linear, logistic, poisson

Value

a list of summary

Examples

x <- matrix(rnorm(1000), ncol = 10);
y <- rowSums(x[,1:5]) + rnorm(100)
m.s <- GibbsSampler(y,x,info = "BIC", family = "gaussian")

Kaukol/IBGS documentation built on June 17, 2024, 2:37 p.m.