We-23: Teaching Bootstrapping Methods with R programming


By Xuemao Zhang (East Stroudsburg University)


Abstract

Bootstrapping is a resampling method used for point estimation, estimating the standard errors of point estimators and confidence interval estimations. The method is generally useful for estimating the distribution of a statistic when normality assumption is not valid. It resamples from the original dataset with replacement of the same size. The boot package in R provides extensive facilities for bootstrapping methods. However, it is beneficial for students to learn the method using R programming due to its resampling nature. In this presentation, I will explain how I taught students the bootstrapping method using the R functions sample and boot, and user-defined functions. The examples include point and interval estimation of a population mean, estimating the coefficients in linear regression models, and estimating accuracy of a linear regression model.


Recording

We-23 - Teaching Bootstrapping Methods with R programming.pdf

register