Resample clusters from an observed data set

bootstrap_clusters(clusters, n_boot = NULL)

Arguments

clusters	data frame of observed data with the following columns cluster_id cluster ID other covariates any other covariates
n_boot	number of clusters to resample. If NULL, will re-sample as many in original cluster

Value

data frame with new cluster_id variable but can be traced back to original cluster through the column original_id

Examples

  data <- data.frame(cluster_id = c("a", "a", "a",
"b", "b",
"c", "c",
"d"),
x = c(1, 0, 1,
     1, 1,
     0, 1,
     0))
boot_clusters <- bootstrap_clusters(clusters = data)
boot_clusters
#>   cluster_id x original_id
#> 1          1 1           a
#> 2          1 0           a
#> 3          1 1           a
#> 4          2 0           c
#> 5          2 1           c
#> 6          3 0           d
#> 7          4 0           c
#> 8          4 1           c