PoSI: Valid Post-Selection Inference for Linear LS Regression

In linear LS regression, calculate for a given design matrix the multiplier K of coefficient standard errors such that the confidence intervals [b - K*SE(b), b + K*SE(b)] have a guaranteed coverage probability for all coefficient estimates b in any submodels after performing arbitrary model selection.

Version: 1.0
Suggests: MASS
Published: 2017-01-15
Author: Andreas Buja, Kai Zhang
Maintainer: Kai Zhang <zhangk at email.unc.edu>
License: GPL-3
NeedsCompilation: no
CRAN checks: PoSI results

Downloads:

Reference manual: PoSI.pdf
Package source: PoSI_1.0.tar.gz
Windows binaries: r-devel: PoSI_1.0.zip, r-release: PoSI_1.0.zip, r-oldrel: PoSI_1.0.zip
OS X El Capitan binaries: r-release: PoSI_1.0.tgz
OS X Mavericks binaries: r-oldrel: PoSI_1.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=PoSI to link to this page.