fun.testregboot: Compare coverage of bootstrap confidence intervals from...
In kolassa-dev/NonparametricHeuristic: Heuristic Tools for Nonparametric Statistics

fun.testregboot

R Documentation

Compare coverage of bootstrap confidence intervals from regression.

Description

Compare coverage of bootstrap confidence intervals from regression.

Usage

fun.testregboot(dists, regparam, sampsize = 10, mcsamp = 10000, B = 9999)

Arguments

`dists`	vector of names of random number generators for sample. Defaults to normal.
`regparam`	function to pass to the bootstrapping tool
`sampsize`	Number of observations per sample; defaults to 10
`mcsamp`	Number of Monte Carlo samples.
`B`	Bootstrap size

Value

A matrix of coverage levels. Dimensions are distribution and confidence interval technique.

Examples

regparam<-function(data,indices,fittedv,residuals){
   y<-fittedv+residuals[indices]
   return(summary(lm(y~data[,1]))$coefficients[2,1:2]^(1:2))}
fun.testregboot(c("rnorm","rcauchy","rexp"),regparam,mcsamp=20,B=99)

kolassa-dev/NonparametricHeuristic documentation built on Nov. 14, 2022, 11:25 p.m.