Premium
Statistical analysis of change in networks
Author(s) -
Frank O.
Publication year - 1991
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/j.1467-9574.1991.tb01310.x
Subject(s) - random graph , graph , computer science , change detection , logit , linear regression , mathematics , logistic regression , econometrics , statistics , theoretical computer science , artificial intelligence
A survey is given of random graphs and random graph processes which can be used to describe and analyze networks that are changing with time. Marko‐vian change over time, log‐linear models for change, and conditionally uniform models for change are described. It is noted that estimation is usually complex if the random graph involves dependent dyads. Models with deterministic change over time may be a way to avoid the difficulties implied by dependent dyads. Logit regression methods are described that can be used to estimate such models.