REBayes (version 0.63)

WGVmix: WGVmix: Weighted Generalized Maximum Likelihood for Empirical Bayes Estimation of Gamma Variances

Description

A Kiefer-Wolfowitz procedure for ML estimation of a Gaussian model with independent variance components with weighted longitudinal data.

Usage

WGVmix(y, id, w, v, pv = 300, eps = 1e-06, rtol = 1e-06, verb = 0, control = NULL)

Arguments

y

A vector of observations

id

A strata indicator vector of the same length

w

A vector of weights

v

A vector of bin boundaries for the variance effects

pv

The number of variance effect bins, if u is missing

eps

A tolerance for determining the support of the bins

rtol

A tolerance for determining duality gap convergence tolerance in Mosek

verb

A flag indicating how verbose the Mosek output should be

control

Mosek control list see KWDual documentation

Value

An object of class density consisting of the following components: components:

Details

See Gu and Koenker (2012?)

References

Gu Y. and R. Koenker (2012) Empirical Bayesball