bootstrap: Calculate bootstrap confidence intervals for XGBP

In meirelesff/xgbp: Flexible and Fast Multilevel Regression with Poststratification

Calculate bootstrap confidence intervals for XGBP

Description

bootstrap uses sample with replacement from provided survey data to calculate non-parametric confidence intervals for the xgbp function

Usage

bootstrap(xgbp_out, boot_iter = 100, ci_level = 0.95, verbose = TRUE)

Arguments

xgbp_out	A xgbp object returned by the xgbp function
boot_iter	Number of bootstrap iterations used to estimate non-parametric confidence intervals. Defaults to 100
ci_level	Level of confidence intervals. Defaults to 0.95 (95% CI)
verbose	Should the function report messages along the estimation? Defaults to TRUE

Parallelization

get_estimates uses furrr's furrr::future_map() to perform bootstrap iterations. By default, computation is done sequentially, but users can take advantage of parallelism by declaring a plan using future::plan(). See below for an example of setting multisession parallelism.

meirelesff/xgbp documentation built on Sept. 24, 2022, 1:48 p.m.