Bootstrapped data sets with replacement — bootstrap

For a user provided data set a number of bootstrapped samples are generated with replacement.

Usage

bootstrap_data(data, times, include_original = FALSE, seed = NULL, ...)

Arguments

data: A data frame from which to generate bootstrap samples.
times: A numeric input specifying how many bootstrap draws are to be generated from data.
include_original: A logical input. If TRUE the returned output will contain a row with the original data. Default is FALSE (see details).
seed: A numeric input specifying the seed to be used. Default is NULL
...: These dots are for future extensions and must be empty.

Value

A nested tibble

Details

Note that by design each bootstrapped sample is of the same size as the original data set. The function generates a simple random sample with replacement along the number of rows in the provided data. The suitability of making any inference from the bootstrapped samples depend on a variety of points (eg. size of original N, number of boot strap draws generated) which are left to the user to decide if they are warranted. The include_original argument can be useful to obtain apparent selection from a given variable selection method.

Examples


test_dat <- data.frame(a = 1:5, b = rnorm(5))
bootstrap_data(test_dat, times = 2, seed = 123) 
#> # A tibble: 2 × 2
#>   boot_rep data        
#>      <int> <list>      
#> 1        1 <df [5 × 2]>
#> 2        2 <df [5 × 2]>