R: Extract parameter (estimator) covariance matrix from GAM fit

vcov.gam {mgcv}

R Documentation

Extract parameter (estimator) covariance matrix from GAM fit

Description

Extracts the Bayesian posterior covariance matrix of the parameters or frequentist covariance matrix of the parameter estimators from a fitted gam object.

Usage

## S3 method for class 'gam'
vcov(object, sandwich=FALSE, freq = FALSE, dispersion = NULL,unconditional=FALSE, ...)

Arguments

`object`	fitted model object of class `gam` as produced by `gam()`.
`sandwich`	compute sandwich estimate of covariance matrix. Currently expensive for discrete bam fits.
`freq`	`TRUE` to return the frequentist covariance matrix of the parameter estimators, `FALSE` to return the Bayesian posterior covariance matrix of the parameters. The latter option includes the expected squared bias according to the Bayesian smoothing prior.
`dispersion`	a value for the dispersion parameter: not normally used.
`unconditional`	if `TRUE` (and `freq==FALSE`) then the Bayesian smoothing parameter uncertainty corrected covariance matrix is returned, if available.
`...`	other arguments, currently ignored.

Details

Basically, just extracts object$Ve, object$Vp or object$Vc (if available) from a gamObject, unless sandwich==TRUE in which case the sandwich estimate is computed (with or without the squared bias component).

Value

A matrix corresponding to the estimated frequentist covariance matrix of the model parameter estimators/coefficients, or the estimated posterior covariance matrix of the parameters, depending on the argument freq.

Author(s)

Henric Nilsson. Maintained by Simon N. Wood simon.wood@r-project.org

References

Wood, S.N. (2017) Generalized Additive Models: An Introductio with R (2nd ed) CRC Press

Examples

 
require(mgcv)
n <- 100
x <- runif(n)
y <- sin(x*2*pi) + rnorm(n)*.2
mod <- gam(y~s(x,bs="cc",k=10),knots=list(x=seq(0,1,length=10)))
diag(vcov(mod))

[Package mgcv version 1.9-1 Index]