This function defines the cluster variables, then resamples from the unique clusters N times, where N is the number of unique clusters. Observations within each cluster are also resampled.
Arguments
- dat
The original data to be bootstrapped (does not contain a variable named "cluster").
- cluster.vars
An unquoted vector of column names used to construct a cluster. An unquoted vector of column names.
- out.folder
A quoted string that gives the path to an existing folder where bootstrapped data objects should be written.
- i
a counter that will be written to the name of each bootstrapped data (e.g bootdat_1.rds, dat_2.rds)