A bootstrap for complex sampling designs

The goal of the surveybootstrap package is to help people use the bootstrap to estimate sampling uncertainty from surveys, including surveys with complex sample designs.

For more information on the bootstrap for surveys, see

• Rao, JNK and Wu, CFJ (1988). Resampling inference with complex survey data. Journal of the American Statistical Association, 83(401):231–241.
• Rao, J., Wu, C., and Yue, K. (1992). Some recent work on resampling methods for complex surveys. Survey Methodology, 18(2):209–217.
• Rust, K. and Rao, J. (1996). Variance estimation for complex surveys using replication techniques. Statistical Methods in Medical Research, 5(3):283-310.
• Efron, B. and Tibshirani, R. J. (1993). An introduction to the bootstrap. Chapman and Hall/CRC.

The development of this software was supported by a grant from the National Institutes of Health (R01-HD075666).

Installing

You can install:

• the latest released version from CRAN with

  install.packages("surveybootstrap")
  

• the latest development version from github with

  install.packages("devtools")
  # devtools::install_github("dfeehan/surveybootstrap")
  install.packages("surveybootstrap", repos = c('https://dfeehan.r-universe.dev', 'https://cloud.r-project.org'))
  

Issues

If you would like to suggest a feature or report a bug, please create an issue

Citation

If you use our package for your research, please cite it so that we can continue to develop it.

• Feehan, Dennis M. and Salganik, Matthew J. (2016) "The surveybootstrap package." http://cran.r-project.org/package=surveybootstrap

Wishlist

• vignette to cover jackknife-after-bootstrap for estimating variance of variance estimators

• non-vignette article with larger-scale demonstration (ie, a demo that takes longer to run than would be good for a vignette)

• either remove the RDS bootstrap or include a vignette to document it

• update to use the purrr package, instead of plyr (but we'll probably need to wait until purrr supports parallelization)

• there are two functions: rescaled.bootstrap.weights and get.rescaled.bootstrap.weights; only need the first one - cleanly remove the second