netmatchRI

netmatchRI provides network-constrained matching and randomization-based inference for observational network data. Dual-penalty matching minimizes total covariate dissimilarity while enforcing penalties on network proximity between and within treatment arms in matched sets. Randomization-based inference combines evidence from correlated matched sets based on this design.

Installation

Install the source archive supplied with the release:

install.packages("netmatchRI_0.1.0.tar.gz", repos = NULL, type = "source")
library(netmatchRI)

HiGHS is the default open-source solver. Gurobi is optional and requires a separate optimizer, license, and R package; see the official installation guide.

Guide and example

The package includes simulated data and a network. The guide covers matching, diagnostics, randomization-based inference, and sensitivity analysis:

data("netmatch_example", package = "netmatchRI")
vignette("netmatchRI", package = "netmatchRI")

The same workflow is available as a runnable script in inst/examples/basic-workflow.R. After installation, locate it with:

system.file("examples", "basic-workflow.R", package = "netmatchRI")

Citation and support

If you use netmatchRI in publications, please cite the associated manuscript:

Dong Z, Lee Y (2026). “Network-Constrained Matching and Randomization-Based Inference for Causal Effects.” arXiv:2610.04102 [stat.ME]. https://doi.org/10.48550/arXiv.2610.04102

Use citation("netmatchRI") for the installed citation. To report a problem or request a feature, open an issue at https://github.com/zhejiadong/netmatchRI/issues.