baseline {remstats}R Documentation

baseline

Description

Specifies an intercept for the tie-oriented model or the sender activity rate step of the actor-oriented model in the same manner as in lm (see Details).

Details

A baseline effect is automatically specified for the tie-oriented model and the sender activity rate step of the actor-oriented model when the ordinal argument in remstats, tomstats, aomstats is set to FALSE (default) and automatically removed when this argument is set to TRUE. Alternatively, a baseline effect can be explicitly specified by adding '1' to the equation or explicitly removed by adding '-1' to the equation.

The baseline effect refers to the baseline tendency to interact. In the tie-oriented model, the log-inverse of the estimated parameter translates to the average number of observed events per time unit per dyad. In the actor-oriented model, the log-inverse of the estimated parameter translates to the average number of observed events per time unit per actor. The statistic is equal to one for all dyads resp. actors in the riskset at all timepoints.

Examples

reh_tie <- remify::remify(history, model = "tie")
remstats(reh = reh_tie, tie_effects = ~1)

reh_actor <- remify::remify(history, model = "actor")
remstats(reh = reh_actor, sender_effects = ~1)

[Package remstats version 3.2.2 Index]