simpleBootstrap: Simple Bootstrap Function

In rettopnivek/utilityf: Useful Functions for Modeling and Plotting

Description

A function for computing a test statistic over multiple replicates of a set of data. Replicates are created by sampling from the original data with replacement.

Usage

simpleBootstrap(x, T_x = mean, nRep = 1000, ...)

Arguments

x	a numeric vector, or a long-form matrix or data frame.
T_x	a function to compute a desired test statistic (default is the mean).
nRep	the number or replicates to sample.
...	additional parameters for the test statistic function.

Value

Returns a list containing the test statistics for each replicate, along with the test statistic for the original set of data.

Examples

# Vector
x = rnorm( 100 )
bootstrap = simpleBootstrap( x, T_x = median )
quantile( bootstrap$replicates, c( .16, .84 ) )

# Matrix
x = cbind( rlnorm( 100, -2, .5 ), rbeta( 100, 4, 1 ) )
bootstrap = simpleBootstrap( x, T_x = colMeans )
apply( bootstrap$replicates, 1, quantile, prob = c( .16, .84 ) )

rettopnivek/utilityf documentation built on March 1, 2021, 7:05 p.m.