Log-linear modeling is a popular method for the analysis of contingency table data. When the table is sparse, the data can fall on the boundary of the convex support, and we say that "the MLE does not exist" in the sense that some parameters cannot be estimated. However, an extended MLE always exists, and a subset of the original parameters will be estimable. The 'eMLEloglin' package determines which sampling zeros contribute to the non-existence of the MLE. These problematic zero cells can be removed from the contingency table and the model can then be fit (as far as is possible) using the glm() function.

Documentation

Manual: eMLEloglin.pdf
Vignette: User manual

Maintainer: Matthew Friedlander <mattyf5 at hotmail.com>

Author(s): Matthew Friedlander

Install package and any missing dependencies by running this line in your R console:

install.packages("eMLEloglin")

Depends lpSolveAPI
Imports
Suggests
Enhances
Linking to
Reverse
depends
Reverse
imports
Reverse
suggests
Reverse
enhances
Reverse
linking to

Package eMLEloglin
Materials
URL
Task Views
Version 1.0.1
Published 2016-11-30
License GPL (>= 2)
BugReports
SystemRequirements
NeedsCompilation no
Citation
CRAN checks eMLEloglin check results
Package source eMLEloglin_1.0.1.tar.gz