A regularization method for the cumulative link models. The 'smooth-effect-on-response penalty' ('SERP') provides flexible modelling of the ordinal model by enabling the smooth transition from the general cumulative link model to a coarser form of the same model. In other words, as the tuning parameter goes from zero to infinity, the subject-specific effects associated with each variable in the model tend to a unique global effect. The parameter estimates of the general cumulative model are mostly unidentifiable or at least only identifiable within a range of the entire parameter space. Thus, by maximizing a penalized rather than the usual non-penalized log-likelihood, this and other numerical problems common with the general model are to a large extent eliminated. Fitting is via a modified Newton's method. Several standard model performance and descriptive methods are also available. An outline of the penalty implemented here is found in Tutz, G and Gertheiss, J (2016) <doi:10.1177/1471082X16642560>.
Version: | 0.1.8 |
Depends: | R (≥ 3.2.0) |
Imports: | ordinal (≥ 2016-12-12), stats |
Suggests: | covr, testthat, VGAM (≥ 1.1-4) |
Published: | 2021-01-29 |
Author: | Ejike R. Ugba [aut, cre, cph] |
Maintainer: | Ejike R. Ugba <ejike.ugba at outlook.com> |
BugReports: | https://github.com/ejikeugba/serp/issues |
License: | GPL-2 |
URL: | https://github.com/ejikeugba/serp |
NeedsCompilation: | no |
Materials: | README NEWS |
CRAN checks: | serp results |
Reference manual: | serp.pdf |
Package source: | serp_0.1.8.tar.gz |
Windows binaries: | r-devel: serp_0.1.8.zip, r-release: serp_0.1.8.zip, r-oldrel: serp_0.1.8.zip |
macOS binaries: | r-release: serp_0.1.8.tgz, r-oldrel: serp_0.1.8.tgz |
Please use the canonical form https://CRAN.R-project.org/package=serp to link to this page.