Change we can believe in: comparing longitudinal network models on consistency, interpretability and predictive power

While several models for analysing longitudinal network data have been proposed, their main differ-ences, especially regarding the treatment of time, have not been discussed extensively in the literature.However, differences in treatment of time strongly impact the conclusions that can be drawn from data.In this article we compare auto-regressive network models using the example of TERGMs – a temporalextensions of ERGMs – and process-based models using SAOMs as an example. We conclude that theTERGM has, in contrast to the ERGM, no consistent interpretation on tie-level probabilities, as well as noconsistent interpretation on processes of network change. Further, parameters in the TERGM are stronglydependent on the interval length between two time-points. Neither limitation is true for process-basednetwork models such as the SAOM. Finally, both compared models perform poorly in out-of-sampleprediction compared to trivial predictive models.

Publication year:
In: Social Networks. - Vol. 52(2018), January, p. 180-191

Note: The status of this file is: restricted

 Record created 2018-02-22, last modified 2019-08-05

Download fulltext

Rate this document:

Rate this document:
(Not yet reviewed)