vcovAcomp {compositions}

Variance covariance matrix of parameters in compositional regression

Description

The variance covariance tensor structured according of linear models with ilr(acomp(...)) responses.

Usage

vcovAcomp(object,...)
          

Arguments

object

a statistical model

...

further optional parameters for vcov

Details

The prediction error in compositional linear regression models is a complicated object. The function should help to organize it.

Value

An array with 4 dimensions. The first 2 are the index dimensions of the ilr transform. The later 2 are the index of the parameter.

Author(s)

K.Gerald v.d. Boogaart http://www.stat.boogaart.de

See Also

vcov

Examples

data(SimulatedAmounts)
model <- lm(ilr(sa.groups)~sa.groups.area)
vcovAcomp(model)[,,1,1]