threeboost: Thresholded variable selection and prediction based on estimating equations

This package implements a thresholded version of the EEBoost algorithm described in [Wolfson (2011, JASA)]. EEBoost is a general-purpose method for variable selection which can be applied whenever inference would be based on an estimating equation. The package currently implements variable selection based on the Generalized Estimating Equations, but can also accommodate user-provided estimating functions. Thresholded EEBoost is a generalization which allows multiple variables to enter the model at each boosting step.

Author
Julian Wolfson and Christopher Miller
Date of publication
2014-08-11 00:18:02
Maintainer
Julian Wolfson <julianw@umn.edu>
License
GPL-3
Version
1.1

View on CRAN

Man pages

coef_traceplot
Draw a coefficient traceplot
eeboost
EEBoost
ee.GEE
GEE estimating functions
geeboost
GEEBoost
QIC
Pan's QIC
threeboost
Thresholded EEBoost
threeboost-package
Thresholded boosting based on estimating equations

Files in this package

threeboost
threeboost/NAMESPACE
threeboost/R
threeboost/R/geefns.R
threeboost/R/eeboost.R
threeboost/R/geeboost.R
threeboost/R/threeboost-package.R
threeboost/R/threeboost.R
threeboost/R/QIC.R
threeboost/R/traceplot.R
threeboost/MD5
threeboost/DESCRIPTION
threeboost/man
threeboost/man/coef_traceplot.Rd
threeboost/man/eeboost.Rd
threeboost/man/threeboost.Rd
threeboost/man/threeboost-package.Rd
threeboost/man/QIC.Rd
threeboost/man/ee.GEE.Rd
threeboost/man/geeboost.Rd