Skip to content
Smoothed bootstrap and functions for sampling from kernel densities
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
R
man
src
tests
.Rbuildignore
.gitignore
.travis.yml
DESCRIPTION
NAMESPACE
README.md
appveyor.yml
cran-comments.md
kernelboot.Rproj

README.md

kernelboot

CRAN_Status_Badge Travis-CI Build Status AppVeyor Build Status Coverage Status Downloads

This package implements random generation procedures for sampling from kernel densities and smoothed bootstrap, that is an extension of standard bootstrap procedure, where instead of drawing samples with replacement from the empirical distribution, they are drawn from kernel density estimate of the distribution.

Three functions are provided to sample from univariate kernel densities (ruvk), multivariate product kernel densities (rmvk) and multivariate Gaussian kernel densities (rmvg). The ruvk function samples from the kernel densities as estimated using the base R density function. It offers possibility of sampling from kernel densities with Gaussian, Epanechnikov, rectangular, triangular, biweight, cosine, and optcosine kernels. The rmvk offers sampling from a multivariate kernel density constructed from independent univariate kernel densities. It is also possible to sample from multivariate Gaussian kernel density using the rmvg function, that allows for correlation between the variables.

Smooth bootstrap is possible by using the kernelboot function, that draws with replacement samples from the empirical distribution, enhances them using noise drawn from the kernel density and evaluates the user-provided statistic on the samples. This procedure can be thought as an extension of the basic bootstrap procedure.

You can’t perform that action at this time.