summary.BayesSUR {BayesSUR}R Documentation

summary method for class BayesSUR

Description

Summary method for class BayesSUR. It includes the argument matching information, Top predictors/responses on average mPIP across all responses/predictors, elpd estimates, MCMC specification, model specification and hyper-parameters. The summarized number of the selected variable corresponds to the posterior mean of the latent indicator variable thresholding at 0.5 by default.

Usage

## S3 method for class 'BayesSUR'
summary(object, Pmax = 0.5, ...)

Arguments

object

an object of class BayesSUR

Pmax

threshold that truncates the estimated coefficients based on thresholding the estimated latent indicator variable. Default is 0.5

...

other arguments

Value

Return a result summary from an object of class BayesSUR, including the CPOs, number of selected predictors with mPIP>Pmax, top 10 predictors on average mPIP across all responses, top 10 responses on average mPIP across all predictors, Expected log pointwise predictive density (elpd) estimates, MCMC specification, model specification (i.e., covariance prior and gamma prior) and hyper-parameters.

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 = 100, burnin = 50, nChains = 2, gammaPrior = "hotspot",
                hyperpar = hyperpar, tmpFolder = "tmp/", output_CPO=TRUE)

## check output
# show the summary information
summary(fit)


[Package BayesSUR version 2.0-1 Index]