compositions (version 2.0-0)

varmlm: Residual variance of a model

Description

Computes the unbiased estimate for the variance of the residuals of a model.

Usage

# S3 method for mlm
var(x,...) 
# S3 method for lm
var(x,...)

Arguments

x

a linear model object

…

Unused, for generic purposes only.

Value

var.lm

returns a scalar giving the estimated variance of the residuals

var.mlm

returns a the estimated variance covariance matrix of the residuals

Details

The difference of this command to var(resid(X)) is that this command correctly adjusts for the degrees of freedom of the model.

See Also

vcov

Examples

# NOT RUN {
data(Orange)
var(lm(circumference~age,data=Orange))
var(lm(cbind(circumference,age)~age,data=Orange))

# }