What nwrem is, in plain words

The idea

When data come as a list of events (who sent what to whom, and when), a relational event model looks at each event and asks: of all pairs of actors who could have interacted at that moment, why was it this pair? The answer is a set of effects that make a pair more likely: the sender is very active, the pair interacted before, the receiver just wrote to the sender, the two are in the same group. nwrem estimates how strong each effect is, in the way a Cox model compares the case that happened with those that could have.

Questions it answers

  • Do people return to earlier partners (inertia)?
  • Is a message more likely to be answered by its receiver (reciprocity)?
  • Do covariates of the actors or the pair change who interacts with whom?

What you need

A list of timestamped events (sender, receiver, time), the actors, and optional actor and pair variables.

What you get

Coefficients with standard errors, likelihood and information criteria, and model comparison.

Use it when

You have a time-ordered list of events and want to know which effects make the observed pair more likely than the alternatives.

Use something else when

  • you want to treat who acts and whom they choose as two separate decisions: use nwdynam.
  • you have waves of a network, not events: use nwsaom.

It was written from Butts (2008) and cross-checked against the R package relevent. nwrem is part of the nwcommands family, and every package here needs the core.

Next

Install nwrem, then the first tutorial. The home page of the package has the technical description.


Back to top

nwcommands is free software under the GNU General Public License, version 3, as are the model packages nwsaom, nwergm, nwdynam, nwrem, nwalaam, nwlsm and nwsbm. Citing and licenses · Get help · GitHub. Issues with the core go to the nwcommands repository; issues with a model package go to that package's own repository.

This site uses Just the Docs, a documentation theme for Jekyll.