|
Activity Number:
|
41
- Statistical Analysis of Networks
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 30, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #324267
|
|
|
Title:
|
Composite Likelihood Estimation for Random-Effect Network Models
|
|
Author(s):
|
Yanjun He* and Peter Hoff
|
|
Companies:
|
University of Washington and Duke University
|
|
Keywords:
|
Random-effect network models ;
Triad ;
Composite Likelihood
|
|
Abstract:
|
Maximum likelihood estimation for random-effect network models involves intractable integrals, while Bayesian approaches using MCMC do not scale well with the number of nodes. As an alternative to these approaches, we propose a composite marginal likelihood estimation method that works with a pseudo-likelihood based on triads of nodes. This approach provides parameter estimates and standard errors at a computational cost that is essentially constant in the number of nodes.
|
Authors who are presenting talks have a * after their name.
