Bootstrap procedure for stepwise regression

Runs bootstrap on models selection procedure using RBIC to find bootstrapped standard error (smoothed, see Efron 2014) as well as selection percentage across candidate variables. (experimental)

Usage

sparseRBIC_bootstrap(srbic_fit, B = 100, quiet = FALSE)

Arguments

srbic_fit

An object fitted by sparseRBIC_step

B

Number of bootstrap samples

quiet

Should the display of a progress bar be silenced?

Value

a list containing:

results

a tibble containing coefficients, p-values, selection pct

bootstraps

a tibble of bootstrapped coefficients