S. Goodreau, and M. Handcock. 47. University of Washington, Center for Statistics and the Social Sciences, Seatle, Washington, USA, (April 2005)
Abstract
We present a systematic examination of real network datasets using maximum likelihood estima-
tion for exponential random graph models as well as new procedures to evaluate how well the
models fit the observed graphs. These procedures compare structural statistics of the observed
graph with the corresponding statistics on graphs simulated from the fitted model. We apply this
approach to the study of friendship relations among high school students from the National Lon-
gitudinal Study of Adolescent Health (AddHealth). The sizes of the networks we fit range from
71 to 2209 nodes. The larger networks represent more than an order of magnitude increase over
the size of any network previously fit using maximum likelihood methods for models of this kind. We argue that several well-studied models in the networks literature do not fit these data well, and
we demonstrate that the fit improves dramatically when the models include the recently-developed
geometrically weighted edgewise shared partner (GWESP), geometrically weighted dyadic shared
partner (GWDSP), and geometrically weighted degree (GWD) network statistics. We conclude
that these models capture aspects of the social structure of adolescent friendship relations not rep-
resented by previous models.
%0 Report
%1 goodreau_goodness_2005
%A Goodreau, Steven M.
%A Handcock, Mark S.
%C Seatle, Washington, USA
%D 2005
%K Model Transitivity checking, clustering, exponential graph model model, networks, random simulations, social verification,
%N 47
%P 22
%T Goodness of fit of social network models
%U http://hbanaszak.mjr.uw.edu.pl/TempTxt/tr0502.pdf
%X We present a systematic examination of real network datasets using maximum likelihood estima-
tion for exponential random graph models as well as new procedures to evaluate how well the
models fit the observed graphs. These procedures compare structural statistics of the observed
graph with the corresponding statistics on graphs simulated from the fitted model. We apply this
approach to the study of friendship relations among high school students from the National Lon-
gitudinal Study of Adolescent Health (AddHealth). The sizes of the networks we fit range from
71 to 2209 nodes. The larger networks represent more than an order of magnitude increase over
the size of any network previously fit using maximum likelihood methods for models of this kind. We argue that several well-studied models in the networks literature do not fit these data well, and
we demonstrate that the fit improves dramatically when the models include the recently-developed
geometrically weighted edgewise shared partner (GWESP), geometrically weighted dyadic shared
partner (GWDSP), and geometrically weighted degree (GWD) network statistics. We conclude
that these models capture aspects of the social structure of adolescent friendship relations not rep-
resented by previous models.
@techreport{goodreau_goodness_2005,
abstract = {We present a systematic examination of real network datasets using maximum likelihood estima-
tion for exponential random graph models as well as new procedures to evaluate how well the
models fit the observed graphs. These procedures compare structural statistics of the observed
graph with the corresponding statistics on graphs simulated from the fitted model. We apply this
approach to the study of friendship relations among high school students from the National Lon-
gitudinal Study of Adolescent Health (AddHealth). The sizes of the networks we fit range from
71 to 2209 nodes. The larger networks represent more than an order of magnitude increase over
the size of any network previously fit using maximum likelihood methods for models of this kind. We argue that several well-studied models in the networks literature do not fit these data well, and
we demonstrate that the fit improves dramatically when the models include the recently-developed
geometrically weighted edgewise shared partner (GWESP), geometrically weighted dyadic shared
partner (GWDSP), and geometrically weighted degree (GWD) network statistics. We conclude
that these models capture aspects of the social structure of adolescent friendship relations not rep-
resented by previous models.},
added-at = {2017-01-09T13:57:26.000+0100},
address = {Seatle, Washington, USA},
author = {Goodreau, Steven M. and Handcock, Mark S.},
biburl = {https://www.bibsonomy.org/bibtex/26a076c13bc3cbd3f3feb32368da2fff6/yourwelcome},
institution = {University of Washington, Center for Statistics and the Social Sciences},
interhash = {21c81440e50c9a65f307b977635b35ba},
intrahash = {6a076c13bc3cbd3f3feb32368da2fff6},
keywords = {Model Transitivity checking, clustering, exponential graph model model, networks, random simulations, social verification,},
month = apr,
number = 47,
pages = 22,
timestamp = {2017-01-09T14:01:11.000+0100},
title = {Goodness of fit of social network models},
url = {http://hbanaszak.mjr.uw.edu.pl/TempTxt/tr0502.pdf},
urldate = {2013-08-11},
year = 2005
}