gfit: Fit an empirical Bayes prior in the hierarchical model mu ~...

In swager/randomForestCI: Confidence Intervals for Random Forests

Description

Fit an empirical Bayes prior in the hierarchical model mu ~ G, X ~ N(mu, sigma^2)

Usage

gfit(X, sigma, p = 2, nbin = 1000, unif.fraction = 0.1)

Arguments

X	a vector of observations
sigma	noise estimate
p	tuning parameter – number of parameters used to fit G
nbin	tuning parameter – number of bins used for discrete approximation
unif.fraction	tuning parameter – fraction of G modeled as "slab"

Value

posterior density estimate g

References

For more details about "g-estimation", see: B Efron. Two modeling strategies for empirical Bayes estimation. Stat. Sci., 29(2): 285–301, 2014.

swager/randomForestCI documentation built on May 30, 2019, 9:33 p.m.