nwcommands

Network analysis for Stata — over 130 commands for importing, generating, describing, analyzing, visualizing, and statistically modeling networks, alongside Stata’s own regular data commands.

Get started Command reference Tutorials

. nwwebuse florentine, nwclear
. nwplot flomarriage, label(_nwnode)
. nwbetween flomarriage
. sum _between

What’s here

Import / export — edgelists, adjacency matrices, UCINET/Pajek-style formats (nwfromedge, nwimport, nwset, nwexport, …)
Generators — random, small-world, preferential-attachment, lattice, and expression-based networks (nwrandom, nwsmall, nwpref, nwlattice, nwgenerate, …)
Structural analysis — centrality, components, cliques/k-plexes/k-cores, cohesive subgroups, community detection, structural equivalence, QAP, and more
Statistical network models — exponential-family random graph models (nwergm), stochastic actor-oriented models (nwsaom), relational event models (nwrem), and Dynamic Network Actor Models (nwdynam)
Visualization — static plots and network movies (nwplot, nwmovie)
Dialog boxes — around 120 GUI dialogs covering nearly every command, reachable via Stata’s own “Network Analysis” menu, for anyone who prefers point-and-click to typing syntax

Every command works alongside ordinary Stata variables — a network can be loaded as Stata variables when you want to, but this is opt-in, not automatic, so building and analyzing a network never has to spend Stata’s own variable budget unless you ask it to.

Installation

. net from "https://raw.githubusercontent.com/thomas-u-grund/nwcommands/master"
. net install nwcommands
. nwinstall, all

The first two lines install just nwinstall itself and the landing help topics; nwinstall, all then downloads everything else — every command, all help files, and every dialog box — and installs them permanently. Add nwinstall, permanently to also add a “Network Analysis” menu to Stata’s own User menu.

See the Installation & Setup tutorial for a full walkthrough, including options for computers without internet access or admin rights.

Learn

New to nwcommands? Start with the tutorials — short, worked walkthroughs with real Stata output, one topic at a time.
Know what command you need? Jump straight to the command reference.
Curious how it’s built, or what’s planned? See the project’s architecture and roadmap on GitHub.

Performance

Two backends exist for the commands where it matters, chosen automatically and transparently — there is nothing to configure, and results are identical either way: a sparse-native Mata graph representation for the core structural commands, proven at 100,000 nodes / 1,000,000 edges, and compiled native (C) kernels for the handful of operations where Stata’s own interpreter, not the algorithm, is the bottleneck.