lik_normalmix: Likelihood object for normal mixture error distribution

In stephens999/ashr: Methods for Adaptive Shrinkage, using Empirical Bayes

Likelihood object for normal mixture error distribution

Description

Creates a likelihood object for ash for use with normal mixture error distribution

Usage

lik_normalmix(pilik, sdlik)

Arguments

pilik	a k vector of mixture proportions (k is the number of mixture components), or an n*k matrix that the j'th row the is mixture proportions for betahat_j
sdlik	a k vector of component-wise standard deviations, or an n*k matrix that the j'th row the is component-wise standard deviations for betahat_j

Examples

   e = rnorm(100,0,0.8) 
   e[seq(1,100,by=2)] = rnorm(50,0,1.5) # generate e~0.5*N(0,0.8^2)+0.5*N(0,1.5^2)
   betahat = rnorm(100)+e
   ash(betahat, 1, lik=lik_normalmix(c(0.5,0.5),c(0.8,1.5)))

stephens999/ashr documentation built on May 16, 2024, 3:10 a.m.