---output: github_document---<!-- README.md is generated from the source: README.Rmd --># bcaboot<!-- badges: start -->[![R-CMD-check](https://github.com/bnaras/bcaboot/workflows/R-CMD-check/badge.svg)](https://github.com/bnaras/bcaboot/actions)[![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/bcaboot)](https://cran.r-project.org/package=bcaboot)[![](https://cranlogs.r-pkg.org/badges/bcaboot)](https://cloud.r-project.org/package=bcaboot)<!-- badges: end -->Bootstrap confidence intervals depend on three elements:- the cdf of the bootstrap replications- the bias-correction number which depends on the proportion of  bootstrap estimates that are less than the original estimate- the acceleration number that measures the rate of  change in standard deviation of the estimate as the data changes.The first two of these depend only on the bootstrap distribution, andnot how it is generated: parametrically ornon-parametrically. Therefore, the only difference in a parametric bcaanalysis would lie in the nonparametric estimation of theacceleration, often a negligible error. The package `bcaboot` provides functions to compute bootstrapconfidence intervals in an (almost) automatic fashion. Further detailsmay be found in the paper by Efron and Narasimhan below.## ReferencesEfron, Bradley, and Balasubramanian Narasimhan. The AutomaticConstruction of Bootstrap Confidence Intervals. (2018)