plsJSS: Linear mixed model deviance function as it appears in the...

In lme4/lme4pureR: lme4 in pure R

Description

A pure R implementation of the penalized least squares (PLS) approach for computing linear mixed model deviances. The purpose is to clarify how PLS works without having to read through C++ code, and as a sandbox for trying out modifications to PLS.

Usage

plsJSS(X, y, Zt, Lambdat, mapping, weights, offset = numeric(n),
  REML = TRUE, ...)

Arguments

X	fixed effects model matrix
y	response
Zt	transpose of the sparse model matrix for the random effects
Lambdat	upper triangular sparse Cholesky factor of the relative covariance matrix of the random effects
mapping	a function that takes a value of theta and produces the non-zero elements of Lambdat. The structure of Lambdat cannot change, only the numerical values
weights	prior weights
offset	offset
REML	calculate REML deviance?
...	additional arguments

Value

a function that evaluates the deviance or REML criterion

lme4/lme4pureR documentation built on May 21, 2019, 7:34 a.m.