WGVmix: WGVmix: Weighted Generalized Maximum Likelihood for Empirical...

In REBayes: Empirical Bayes Estimation and Inference

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

Details

See Gu and Koenker (2012?)

Value

An object of class density consisting of the following components:

x	the variance bin boundaries
y	the function values of the mixing density for the variances.
logLik	the value of the log likelihood at the solution
status	the mosek convergence status.

Author(s)

R. Koenker

References

Gu Y. and R. Koenker (2017) Empirical Bayesball Remixed: Empirical Bayes Methods for Longitudinal Data, J. of Applied Econometrics, 32, 575-599.

Koenker, R. and J. Gu, (2017) REBayes: An R Package for Empirical Bayes Mixture Methods, Journal of Statistical Software, 82, 1–26.

REBayes documentation built on Aug. 19, 2023, 5:10 p.m.