SQB: Sequential Bagging on Regression

Methodology: Remove one observation. Training the rest of data that are sampled without replacement and given this observation's input, predict the response back. Replicate this N times and for each response, take a sample from these replicates with replacement. Average each responses of sample and again replicate this step N time for each observation. Approximate these N new responses by using bootstrap method and generate another N responses y'. Training these y' and predict to have N responses of each testing observation. The average of N is the final prediction. Each observation will do the same.

Version: 0.4
Depends: R (≥ 3.2.5), rpart , parallel, caret, nnet, pls
Suggests: MASS
Published: 2018-07-07
Author: Moshu Xie
Maintainer: Moshu Xie <mxie622 at gmail.com>
License: GPL-2
NeedsCompilation: no
CRAN checks: SQB results

Downloads:

Reference manual: SQB.pdf
Package source: SQB_0.4.tar.gz
Windows binaries: r-devel: SQB_0.4.zip, r-devel-gcc8: SQB_0.4.zip, r-release: SQB_0.4.zip, r-oldrel: SQB_0.4.zip
OS X binaries: r-release: SQB_0.4.tgz, r-oldrel: SQB_0.4.tgz
Old sources: SQB archive

Linking:

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