`RLRsim`

implements fast simulation-based exact tests for variance components in mixed and additive models for
conditionally Gaussian responses – i.e., tests for questions like:

is the variance of my random intercept significantly different from 0?

is this smooth effect significantly nonlinear?

is this smooth effect significantly different from a constant effect?

The convenience functions `exactRLRT`

and `exactLRT`

can deal with fitted models from packages lme4, nlme, gamm4, SemiPar and
from mgcv's `gamm()`

-function.
Workhorse functions `LRTSim`

and `RLRTSim`

accept design matrices as inputs directly and can thus be used more generally
to generate exact critical values for the corresponding
(restricted) likelihood ratio tests.

The theory behind these tests was first developed in:

Crainiceanu, C. and Ruppert, D. (2004)
Likelihood ratio tests in
linear mixed models with one variance component, *Journal of the Royal
Statistical Society: Series B*, **66**, 165–185.

Power analyses and sensitivity studies for RLRsim can be found in:

Scheipl, F., Greven, S. and Kuechenhoff, H. (2008)
Size and power of tests
for a zero random effect variance or polynomial regression in additive and
linear mixed models. *Computational Statistics and Data Analysis*,
**52**(7), 3283–3299.

Fabian Scheipl ([email protected]), Ben Bolker

Embedding an R snippet on your website

Add the following code to your website.

For more information on customizing the embed code, read Embedding Snippets.