gswdesign-package: gswdesign: The Gram-Schmidt walk design

In fsavje/gswdesign-R: The Gram-Schmidt walk design

Description

Provides wrappers for the Julia package 'GSWDesign', which is a fast implementation of the Gram-Schmidt Walk for balancing covariates in randomized experiments. The Gram-Schmidt Walk design allows experimenters to control the trade-off between the amount of covariate balance and robustness provided by treatment assignment.

Details

See sample_gs_walk for the main function.

See the packages' websites for more information: https://github.com/crharshaw/GSWDesign.jl, https://github.com/fsavje/gswdesign-R.

Bug reports and suggestions are greatly appreciated. They are best reported here: https://github.com/fsavje/gswdesign-R/issues.

References

Harshaw, Christopher and Fredrik Sävje and Daniel Spielman and Peng Zhang (2020), ‘Balancing covariates in randomized experiments using the Gram-Schmidt walk’, arXiv 1911.03071. https://arxiv.org/abs/1911.03071

fsavje/gswdesign-R documentation built on Aug. 9, 2020, 7:48 p.m.