Finds the largest possible regression model that will still converge
for various types of regression analyses (including mixed models and generalized
additive models) and then optionally performs stepwise elimination similar to the
forward and backward effect-selection methods in SAS, based on the change in
log-likelihood or its significance, Akaike's Information Criterion, the Bayesian
Information Criterion, the explained deviance, or the F-test of the change in R².
Version: |
1.9 |
Depends: |
R (≥ 3.2) |
Imports: |
graphics, lme4, methods, mgcv, nlme, plyr, stats, utils |
Suggests: |
GLMMadaptive, MASS, gamm4, glmertree, glmmTMB, knitr, lmerTest, nnet, ordinal, parallel, partykit, pbkrtest, rmarkdown, testthat |
Published: |
2021-03-27 |
Author: |
Cesko C. Voeten
[aut, cre] |
Maintainer: |
Cesko C. Voeten <cvoeten at gmail.com> |
BugReports: |
https://github.com/cvoeten/buildmer/issues |
License: |
FreeBSD |
NeedsCompilation: |
no |
Materials: |
ChangeLog |
CRAN checks: |
buildmer results |