glmmRTMB/glmmTMB/man/sandwich.Rd at main · glmmTMB/glmmRTMB

Repository navigation

% Generated by roxygen2: do not edit by hand

% Please edit documentation in R/methods.R

\name{sandwich}

\alias{sandwich}

\alias{sandwich.default}

\alias{sandwich.glmmTMB}

\title{Sandwich Estimator based on Bread and Meat Matrices}

\usage{

sandwich(x, ...)

\method{sandwich}{default}(x, ...)

\method{sandwich}{glmmTMB}(x, full = FALSE, cluster = getGroups(x), rawnames = FALSE, ...)

}

\arguments{

\item{x}{a \code{glmmTMB} object fitted with ML (REML is not supported).}

\item{...}{ignored by the \code{glmmTMB} method.}

\item{full}{logical; if \code{TRUE}, return the full sandwich matrix including variance components,

otherwise only the fixed effects part (if the model was fit with ML).}

\item{cluster}{a factor indicating the cluster structure of the data.}

\item{rawnames}{logical; if \code{TRUE}, keep the original names of the parameters as in the

\code{TMB} object. By default, \code{FALSE} such that the names are sanitized to user friendly names.}

}

\value{

A square matrix representing the sandwich estimator.

}

\description{

This (simplified) method for a new S3 generic based on \code{\link[sandwich]{sandwich}}

computes the sandwich estimator for a fitted \code{glmmTMB} model.

}

\examples{

m <- glmmTMB(count ~ mined + (1 | site), data = Salamanders, family = nbinom1)

sandwich(m)

sandwich(m, full = TRUE)

}