Learn R Programming

natstrat

Natural strata can be used in observational studies to balance the distributions of many covariates across any number of treatment groups and any number of comparisons. These strata have proportional amounts of units within each stratum across the treatments, allowing for simple interpretation and aggregation across strata. Within each stratum, the units are chosen using randomized rounding of a linear program that balances many covariates.

Installation

You can install the released version from CRAN with:

install.packages("natstrat")

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("kkbrum/natstrat", build_vignettes = TRUE)

Usage

To learn about how to use this package, please see the associated vignette with:

browseVignettes(package = "natstrat")

Copy Link

Version

Install

install.packages('natstrat')

Monthly Downloads

78

Version

2.0.1

License

GPL-3

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/kkbrum/natstrat

Homepage

https://kkbrum.github.io/natstrat/

Maintainer

Katherine Brumberg

Last Published

February 3rd, 2026

Functions in natstrat (2.0.1)

verify_inputs_EMD

Verify the inputs to the earthmover's distance problem

verify_multi_comp_inputs

Verify the inputs for supplemental comparisons to optimize_controls()

create_balance_matrices

Create matrix of balance constraints for linear program

Calculate desired number of controls per stratum

create_dist_matrix

Create matrix of distances between strata

generate_constraints

Generate constraints to encourage covariate balance

Check covariate balance of the control and treated groups

optimize_controls

Select control units that optimize covariate balance

Standardize covariate vector for balance constraint

plot_stand_diffs

Plot standardized differences in means

get_stand_diffs

Calculate standardized differences

Linear program that selects which controls to use in order to optimize balance

Homocysteine and smoking example data with multiple control groups

Homocysteine and smoking example data

Solve the earthmover's distance problem

randomized_rounding_expectation

Sample integer solution from linear programming solution with sample sizes correct in expectation

Parse the nonstandard balance formulas

natstrat-package

natstrat: Obtain Unweighted Natural Strata that Balance Many Covariates

Verify the inputs to optimize_controls()

Process the desired sample sizes for optimize_controls()

randomized_rounding

Sample integer solution from linear programming solution with correct sample sizes