Effects

An effect is one rule in the model, with one coefficient. In nwsaom you request an effect by naming it as an option. This page is only a map; the complete list, with RSiena’s names, is in Stata’s help nwsaom (and help nwsaom_remarks) and in the manual (PDF) in the GitHub repository. detail after a fit prints the RSiena name of each effect.

Family	Examples (options)	Question
Baseline	outdegree, reciprocity	density; return of ties
Triadic closure	transtrip, transties, transmedtrip, transrectrip, cycle3, cycle4, balance, gwesp()	clustering of friends of friends
Degree effects	indegpopularity, outpopularity, outactivity, inactivity, outtrunc, outmore, …	popular and active actors
Isolates and assortativity	isolatenet, outiso, outoutass, ininass, inoutass, …	isolated actors; degree matching
Covariate effects	egox(), altx(), samex(), simx(), absdiffx(), diffx(), higher(), dyadcov()	who sends, who receives, who chooses similar others
Interactions	interact()	does one effect depend on another
Endowment and creation	endow(), creation()	different rules for keeping and for creating ties
Rate effects	ratecov()	who gets more opportunities to change
Behavior	behavior() with linear, quadratic, avalt, avsim, beheff()	influence and the shape of behavior change
Network on behavior	behsim, behegox, behaltx	selection on the behavior
Several networks	nwsaom multinet with eff1(), cross1(), …	effects of one network on another
Two-mode, undirected	detected from the data; symmetric, symtype()	affiliation and mutual ties

Used in the tutorials: 1 (baseline, closure, covariates), 3 (degree effects, gwesp), 4 (behavior effects), 6 (cross-network effects), 8 (changing and dyadic covariates), 9 (undirected and two-mode networks) and 10 (endowment, creation, interactions, rate effects). For how the effects relate to RSiena’s table see nwsaom and RSiena.