Social networks can be defined as the patterns of ties between social actors
as encountered, e.g., when studying friendship between
individuals, cooperation between companies, or trade between countries.
For the statistical analysis of such data, representing the dependence
between ties is a crucial concern. Recent years have seen a surge in the
development of statistical models and procedures for social network
data, and this presentation will give an overview, focusing on
longitudinal (i.e., panel) data.
To represent the feedback processes inherent in network dynamics, it is
helpful to regard such panel data as momentary observations on a
continuous-time process on the space of directed graphs. Tie-oriented and
actor-oriented stochastic models are presented, which can reflect endogenous
network dynamics such as transitivity as
well as effects of actor-level or dyadic covariates.
These models do not allow explicit calculations, but they can be
implemented as computer simulation models. Stochastic approximation
methods can be used to estimate the parameters.
Some examples are given, and some open problems and directiuons for
future work are discussed.
Presented by:
Tom Snijders (University of Groningen)
Date & time:
November 13, 2006 4:00 pm - November 13, 2006 12:00 am
External seminars home