Tightens an Observational Block Design by Balanced Subset Matching

Tightens an observational block design into a smaller design with either smaller or fewer blocks while controlling for covariates. The method uses fine balance, optimal subset matching (Rosenbaum, 2012 ) and two-criteria matching (Zhang et al 2023 ). The main function is tighten(). The suggested 'rrelaxiv' package for solving minimum cost flow problems: (i) derives from Bertsekas and Tseng (1988) , (ii) is not available on CRAN due to its academic license, (iii) may be downloaded from GitHub at < https://github.com/josherrickson/rrelaxiv/>, (iv) is not essential to use the package.