gES: MCMC algorithm for graphical exponential screening estimator
In zhejosephliu/gES: Graphical exponential screening estimator

Description Usage Arguments Value Examples

Function of MCMC algorithm to compute graphical exponential screening estimator.

1	ges.mcmc(x1, x2, hard.thres = 0.05, num.burn.in = 4000, num.estimation = 4000, plot.mcmc = TRUE)

`x1`	An `n` by `p` data matrix used for constructing individual graph estimators, where `n` is the sample size and `p` is the dimension.
`x2`	An `m` by `p` data matrix used for aggregating individual estimators, where `m` is the sample size and `p` is the dimension.
`hard.thres`	Hard thresholding parameter for covariance matrix estimation.
`num.burn.in`	Number of burn-in iterations in MCMC algorithm.
`num.estimation`	Number of MCMC iterations for precision matrix estimation.
`plot.mcmc`	If `TRUE`, plot the number of selected edges as a function of MCMC steps.

Estimated precision matrix via graphical exponential screening.

library(huge)

n <- 400
p <- 50
model <- huge.generator(n = n, d = p, graph = "hub", g = 5, v = 0.3)
x <- model$data
est.ges <- ges.mcmc(x[1:(n / 2), ], x[(n / 2 + 1):n, ])