Installing nwergm
nwergm is a separate package (GPL-3) that needs the core nwcommands. Install the core first, then nwergm.
1. The core
. net install nwcommands, from(https://nwcommands.org/net/) replace
. nwinstall, everything
nwinstall, everything installs the core and all four model packages in one step; nwinstall, all installs the core only.
See Installation & Setup for details, including computers without internet access or administrator rights.
2. nwergm
. nwinstall, models
If you used nwinstall, everything above, the model packages are already installed; after nwinstall, all (core only) add them with nwinstall, models. Repeating either command updates them. To install nwergm only, or a specific tagged version, use its own address:
. net install nwergm, from(https://raw.githubusercontent.com/thomas-u-grund/nwergm/main/)
(nwinstall, modelpkg(nwergm) does the same from the core.)
A specific release
Every release is tagged on GitHub and each tag has its own address. For version 1.3.7:
. net install nwergm, from(https://raw.githubusercontent.com/thomas-u-grund/nwergm/v1.3.7/) replace
Use a tagged release when you want an analysis to be reproducible: you can then say exactly which version produced the results. The operator terms and the other additions described in this section need version 1.1.0 or later.
3. Check the installation
. nwergm version
nwergm version 1.3.7, 06nov2026
core package nwcommands : installed (development copy, no distribution date)
native plugin : .../nwergm/lib/plugins/macos/ergm_mcmc_macos.plugin (ships with
> the package, no separate version number)
compared with : R ergm 4.12.0
license : GNU GPL-3
The output shows the release of nwergm, the core, the native plugin and the version of ergm
that the results were compared with. (The core line shows “development copy” because the examples on
this website were run from a source checkout; a normal installation shows the distribution date of the
core.) The release is stored in r(version), and after a fit in e(version); quote it when you
report an analysis. nwversion lists the core and all installed model packages. help nwergm opens
the complete documentation.
Without internet access
On a computer without internet access, install from zip files. Download the nwergm zip and the nwcommands zip elsewhere, unzip them, rename the unzipped nwergm-main folder to nwergm, and run nwinstall, localcopy all in the core folder followed by nwinstall, modelpkg(nwergm) modelsfrom("<folder that holds nwergm>"). The steps, with all five zip files, are on the installation page of the core.
Updating
Repeat nwinstall, models (or nwinstall, everything), or, for this package only, its net install line with replace:
. net install nwergm, from(https://raw.githubusercontent.com/thomas-u-grund/nwergm/main/) replace
Updating the core never overwrites nwergm, and updating nwergm never changes the core.
net describe nwergm shows the installed version and distribution date.
The native plugin
The Monte Carlo sampler at the heart of a dyad-dependent fit runs in a compiled plugin that nwergm ships for
macOS, Linux and Windows. It is installed together with the package and found automatically; there is
nothing to configure. nwergm version shows the path of the plugin in use. The help file
(help nwergm, section on performance) says what happens on a platform without a plugin.
If the plugin does not work on your computer
If the plugin of your platform is missing, older than the .ado files (the Linux and Windows files are built by an
automatic process and can lag behind the macOS file for a while), or does not load on your system, compile it on your own computer:
. nwinstall, buildplugin(nwergm)
The command fetches the C source of the package from its GitHub repository, compiles it with the compiler of your computer and
installs the plugin where net install puts it (an existing file is kept with the extension .prev). You need a C compiler and
make: on a Mac the command line tools of Xcode (xcode-select --install in the Terminal), on Linux gcc and make. Restart Stata
afterwards. A plugin depends on the operating system and the processor, not on the version or the flavour of Stata (BE, SE, MP), so a plugin built
on your computer works in every Stata installed there. The routine was tested on macOS; on Windows it stops with a message and the
build note that comes with the source gives the command for the Visual Studio command line. The source is free software (GPL-3); net get nwergm, from(https://raw.githubusercontent.com/thomas-u-grund/nwergm/main)
copies it into the current directory, and it is in the folder native/ of the repository.
Rebuilding the Mata libraries on your own computer
The Mata libraries of the packages are built with Stata 17. A library built by an older Stata loads in every later Stata, but a library built by a newer Stata is ignored by an older one (Stata says “compiled by Stata 18, is too new to be run by this version of Stata and so was ignored”, and every Mata function is then “not found”, r(3499)). If that happens, build the libraries with your own Stata:
. nwinstall, buildmata(nwergm)
The command fetches the Mata source from the GitHub repositories, compiles it with the Stata that is running and installs the libraries where Stata finds the current
ones, keeping an existing file as a .prev copy. buildmata() takes nwcommands, nwsaom, nwergm, nwdynam, nwrem or all (the core and every installed package).
Three things to know: the build clears the data and the Mata objects in memory (it stops with error 4 if data is loaded, so save first); the libraries of nwsaom and
nwergm are compiled against the core library, so rebuilding the core rebuilds the installed nwsaom and nwergm libraries as well; and a build takes a few minutes.
Update the packages first (nwinstall, all and nwinstall, models), because the source of the current release is used. Restart Stata afterwards. The shipped
libraries are the better choice when they load; a build with your own Stata is the fallback and the repair.
Requirements
Stata 17 or later (Stata BE is enough; tested with Stata 17 and 18, the Mata libraries are built with Stata 17, the oldest tested version; Stata 16 and older are untested) and the core nwcommands with its Mata library.