plotMCMCdiag {BayesSUR}

plot MCMC diagnostic plots

Description

Show trace plots and diagnostic density plots of a fitted model object of class BayesSUR.

Usage

plotMCMCdiag(x, nbloc = 3, HIWg = NULL, header = "", ...)

Arguments

x

an object of class BayesSUR

nbloc

number of splits for the last half iterations after substracting burn-in length

HIWg

diagnostic plot of the response graph. Default is NULL. HIW="degree" prints the diagnostic of the degrees of response nodes. HIW="edges" prints the diagnostic of every edge between two responses. HIW="lik" prints the diagnostic of the posterior likelihoods of the hyperparameters related to the response relationships

header

the main title

...

other arguments for the plots of the log-likelihood and model size

Examples

data("exampleEQTL", package = "BayesSUR")
hyperpar <- list(a_w = 2, b_w = 5)

set.seed(9173)
fit <- BayesSUR(
  Y = exampleEQTL[["blockList"]][[1]],
  X = exampleEQTL[["blockList"]][[2]],
  data = exampleEQTL[["data"]], outFilePath = tempdir(),
  nIter = 10, burnin = 0, nChains = 1, gammaPrior = "hotspot",
  hyperpar = hyperpar, tmpFolder = "tmp/"
)

## check output
plotMCMCdiag(fit)