WiSEBoot: Wild Scale-Enhanced Bootstrap

Perform the Wild Scale-Enhanced (WiSE) bootstrap. Specifically, the user may supply a single or multiple equally-spaced time series and use the WiSE bootstrap to select a wavelet-smoothed model. Conversely, a pre-selected smooth level may also be specified for the time series. Quantities such as the bootstrap sample of wavelet coefficients, smoothed bootstrap samples, and specific hypothesis testing and confidence region results of the wavelet coefficients may be obtained. Additional functions are available to the user which help format the time series before analysis. This methodology is recommended to aid in model selection and signal extraction. Note: This package specifically uses wavelet bases in the WiSE bootstrap methodology, but the theoretical construct is much more versatile.

Version: 1.3.0
Imports: wavethresh, FAdist
Suggests: knitr
Published: 2015-10-28
Author: Megan Heyman, Snigdhansu Chatterjee
Maintainer: Megan Heyman <heyma029 at umn.edu>
License: GPL-2
NeedsCompilation: no
Materials: NEWS
CRAN checks: WiSEBoot results


Reference manual: WiSEBoot.pdf
Vignettes: WiSEBoot Vignette
Package source: WiSEBoot_1.3.0.tar.gz
Windows binaries: r-devel: WiSEBoot_1.3.0.zip, r-release: WiSEBoot_1.3.0.zip, r-oldrel: WiSEBoot_1.3.0.zip
OS X Snow Leopard binaries: r-release: WiSEBoot_1.3.0.tgz, r-oldrel: not available
OS X Mavericks binaries: r-release: WiSEBoot_1.3.0.tgz
Old sources: WiSEBoot archive