model-coerce {pbkrtest}

R Documentation

Conversion between a model object and a restriction matrix

Description

Testing a small model under a large model corresponds imposing restrictions on the model matrix of the larger model and these restrictions come in the form of a restriction matrix. These functions converts a model to a restriction matrix and vice versa.

Usage

model2remat(largeModel, smallModel, sparse = FALSE)

remat2model(largeModel, L, REML = TRUE, ...)

make_modelmat(X, L)

make_remat(X, X2)

Arguments

largeModel, smallModel	Model objects of the same "type". Possible types are linear mixed effects models and linear models (including generalized linear models)
sparse	Should the restriction matrix be sparse or dense?
L	A restriction matrix; a full rank matrix with as many columns as 'X' has.
REML	Controls if new model object should be fitted with REML or ML.
...	Additional arguments; not used.
X, X2	Model matrices. Must have same numer of rows.

Details

'make_remat' Make a restriction matrix. If span(X2) is in span(X) then the corresponding restriction matrix 'L' is returned.

Value

model2remat: A restriction matrix. remat2model: A model object.

Note

That these functions are visible is a recent addition; minor changes may occur.

Author(s)

Ulrich Halekoh uhalekoh@health.sdu.dk, Søren Højsgaard sorenh@math.aau.dk

References

Ulrich Halekoh, Søren Højsgaard (2014)., A Kenward-Roger Approximation and Parametric Bootstrap Methods for Tests in Linear Mixed Models - The R Package pbkrtest., Journal of Statistical Software, 58(10), 1-30., https://www.jstatsoft.org/v59/i09/

See Also

PBmodcomp, PBrefdist, KRmodcomp

Examples

Run examples

library(pbkrtest)
data("beets", package = "pbkrtest")
sug <- lm(sugpct ~ block + sow + harvest, data=beets)
sug.h <- update(sug, .~. - harvest)
sug.s <- update(sug, .~. - sow)

## Construct restriction matrices from models
L.h <- model2remat(sug, sug.h); L.h
L.s <- model2remat(sug, sug.s); L.s

## Construct submodels from restriction matrices
mod.h <- remat2model(sug, L.h); mod.h
mod.s <- remat2model(sug, L.s); mod.s

## Sanity check: The models have the same fitted values and log likelihood
plot(fitted(mod.h), fitted(sug.h))
plot(fitted(mod.s), fitted(sug.s))
logLik(mod.h)
logLik(sug.h)
logLik(mod.s)
logLik(sug.s)

---