6. Exponential Family Models for Sampled and Census Network Data

Sociological Methodology - Tập 34 Số 1 - Trang 241-270 - 2004
Laura M. Koehly1, Steven M. Goodreau2, Martina Morris2
1Texas A&M University;
2University of Washington

Tóm tắt

Much progress has been made on the development of statistical methods for network analysis in the past ten years, building on the general class of exponential family random graph (ERG) network models first introduced by Holland and Leinhardt (1981). Recent examples include models for Markov graphs, “p*” models, and actor-oriented models. For empirical application, these ERG models take a logistic form, and require the equivalent of a network census: data on all dyads within the network. In a largely separate stream of research, conditional log-linear (CLL) models have been adapted for analyzing locally sampled (“egocentric”) network data. While the general relation between log-linear and logistic models is well known and has been exploited in the case of a priori blockmodels for networks, the relation for the CLL models is different due to the treatment of absent ties. For fully saturated tie independence models, CLL and ERG are equivalent and related via Bayes' rule. For other tie independence models, the two do not yield equivalent predicted values, but we show that in practice the differences are unlikely to be large. The alternate conditioning in the two models sheds light on the relationship between local and complete network data, and the role that models can play in bridging the gap between them.

Từ khóa


Tài liệu tham khảo

10.1002/0471249688

10.1111/j.2517-6161.1974.tb00999.x

10.2307/2987782

Besag J. E., 1977, Bulletin of the International Statistical Association, 47, 77

10.1016/S0022-5193(03)00060-2

Burt R. S., 1983, Applied Network Analysis, 158

10.1016/0378-8733(84)90007-8

Burt R. S., 1990, Structures of Power and Constraint: Papers in Honor of Peter M. Blau

10.4159/9780674029095

10.1097/00002030-200108004-00002

Fienberg S. E., 1977, The Analysis of Cross-Classified Categorical Data

10.2307/270741

Fischer C. S., 1982, To Dwell Among Friends: Personal Networks in Town and City

Frank O., 1988, Mathematiques, Informatique, et Sciences Humaines, 26, 19

Frank O., 1994, Journal of Official Statistics, 10, 53

10.1080/01621459.1986.10478342

Geyer C. J., 1992, Journal of the Royal Statistical Society, 54, 657, 10.1111/j.2517-6161.1992.tb01443.x

Gilks W. R., 1996, Markov Chain Monte Carlo in Practice

Gintis H., 2000, Game Theory Evolving: A Problem-Centered Introduction to Modeling Strategic Interaction

Goodman L., 1984, The Analysis of Cross-classified Data Having Ordered Categories

10.1086/223932

10.1086/225469

Handcock M., 2003, Dynamic Social Network Modeling and Analysis, 229

10.1198/016214502388618906

10.1086/224954

10.2307/270703

10.1080/01621459.1981.10477598

Iacobucci D., 1994, Social Network Analysis: Methods and Applications, 605

10.1080/10618600.1996.10474713

10.2307/2092176

Laumann Edward O., 1973, Bonds of Pluralism: The Form and Substance of Urban Social Networks

10.1126/science.2543079

Laumann E. O., 1987, The Organizational State: Social Change in National Policy Domains

10.1093/biomet/73.1.13

10.1146/annurev.soc.28.110601.141117

10.1080/03050629908434955

10.2307/2095670

10.1016/0378-8733(81)90002-2

10.2307/2095397

10.1016/0378-8733(88)90010-X

10.1177/0002716290510001005

10.1016/0025-5564(91)90014-A

10.1093/0199269017.001.0001

10.1093/oxfordjournals.aje.a117241

10.1086/230190

10.1086/229534

10.1111/1467-8306.9302004

Pattison P. E., 2002, Sociological Methodology, 32, 301

10.1348/000711099159053

10.1177/000312240006500508

10.1001/jama.1997.03550100049038

Rindfuss R., Network Epidemiology: A Handbook For Survey Design and Data Collection

10.1016/S0378-8733(01)00029-6

10.1007/BF02294834

10.1007/BF02294302

10.1080/01621459.1995.10476493

10.1111/0081-1750.00099

10.1080/01621459.1990.10475327

Thompson S. K., 2000, Survey Methodology, 26, 87

10.1017/CBO9780511571138.009

10.1017/CBO9780511815478

10.1007/BF02294547

Watts D. J., 2003, Six Degrees: The Science of a Connected Age

10.1086/229572

10.1111/j.0081-1750.2003.00135.x