est_mixprop_mode: Estimate mixture proportions and mode of the unimodal...

In mengyin/vashr: Variance Adaptive Shrinkage

Description

Estimate mixture proportions and mode of the unimodal inverse-gaama mixture variance prior.

Usage

est_mixprop_mode(sehat, g, prior, df, unimodal, singlecomp, estpriormode,
  maxiter = 5000)

Arguments

sehat	n vector of standard errors of observations
g	the initial prior distribution for variances
prior	numeric vector indicating Dirichlet prior on mixture proportions
df	appropriate degrees of freedom for chi-square distribution of sehat^2
unimodal	put unimodal constraint on the prior distribution of variances ("variance") or precisions ("precision")
singlecomp	logical, indicating whether to use a single inverse-gamma distribution as the prior distribution for the variances
estpriormode	logical, indicating whether to estimate the mode of the unimodal prior
maxiter	maximum number of iterations of the EM algorithm

Value

A list, including the final loglikelihood, the fitted prior g, number of iterations and a flag to indicate convergence.

Examples

fitted.prior = est_mixprop_mode(sehat=abs(rnorm(100)),g=igmix(c(.5,.5),c(1,3),c(1,3)),
prior=c(1,1),df=10,unimodal="variance",singlecomp=FALSE,estpriormode=TRUE)

mengyin/vashr documentation built on May 22, 2019, 6:51 p.m.