sampleGlobalErrorVarianceOneIteration: Sample one iteration of the (inverse gamma) global variance...

In stochtree: Stochastic Tree Ensembles (XBART and BART) for Supervised Learning and Causal Inference

Sample one iteration of the (inverse gamma) global variance model

Description

Sample one iteration of the (inverse gamma) global variance model

Usage

sampleGlobalErrorVarianceOneIteration(residual, dataset, rng, a, b)

Arguments

residual	Outcome class
dataset	ForestDataset class
rng	C++ random number generator
a	Global variance shape parameter
b	Global variance scale parameter

Value

None

Examples

X <- matrix(runif(10*100), ncol = 10)
y <- -5 + 10*(X[,1] > 0.5) + rnorm(100)
y_std <- (y-mean(y))/sd(y)
forest_dataset <- createForestDataset(X)
outcome <- createOutcome(y_std)
rng <- createCppRNG(1234)
a <- 1.0
b <- 1.0
sigma2 <- sampleGlobalErrorVarianceOneIteration(outcome, forest_dataset, rng, a, b)

stochtree documentation built on April 4, 2025, 2:11 a.m.