ctm_package: ctm: Conditional Transformation Models

Description Details Author(s) References

Description

An experimental implementation of conditional transformation models for the semiparametric estimation of conditional distribution functions. Contains example analyses and a simulation study.

Details

Package: ctm
Type: Package
Version: 0.0-3
Date: 2012-09-04
License: GPL-2

This package is highly experimental! DO NOT USE this package for anything beyond demonstration and theoretical experiments.

The package contains a very rough implementation of conditional transformation models as introduced by Hothorn, Kneib, and Buehlmann (2012). This class of semiparametric regression models fits conditional transformation functions that can be used to evaluate the conditional distribution of the response given explanatory variables.

The applications presented in Hothorn, Kneib, and Buehlmann (2013) can be reproduced by running ‘make’ in the inst/applications directory.

The empirical evaluation reported on in the paper can be reproduced by running R on the file sim.R in the directory inst/empeval.

Author(s)

Torsten Hothorn [email protected]

References

Torsten Hothorn, Thomas Kneib, and Peter Buehlmann (2013), Conditional Transformation Models. Journal of the Royal Statistical Society: Series B (Statistical Methodology), accepted 2012-11-27. Preprint available from http://arxiv.org/abs/1201.5786v1.


ctmDevel documentation built on May 2, 2019, 4:52 p.m.