Asymptotic normality of maximum likelihood and its variational approximation for stochastic blockmodels

2012/07/31 by Peter Bickel, David Choi, Xiangyu Chang +1 · 1 citation
Mathematics · Computer Science · #math.ST #cs.SI #stat.TH

paper · pdf · doi:10.1214/13-aos1124

published as Annals of Statistics 2013, Vol. 41, No. 4, 1922-1943 · Published in at http://dx.doi.org/10.1214/13-AOS1124 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

arxiv created 2013/10/29 · arxiv updated 2013/10/30

Abstract

Variational methods for parameter estimation are an active research area, potentially offering computationally tractable heuristics with theoretical performance bounds. We build on recent work that applies such methods to network data, and establish asymptotic normality rates for parameter estimates of stochastic blockmodel data, by either maximum likelihood or variational estimation. The result also applies to various sub-models of the stochastic blockmodel found in the literature.

Cited by

• A Statistical Modelling Approach to Detecting Community in Networks

Related

• Consistency of maximum-likelihood and variational estimators in the Stochastic Block Model
• Optimality of variational inference for stochastic block model with\n missing links
• An Overview of Asymptotic Normality in Stochastic Blockmodels: Cluster Analysis and Inference
• Likelihood-based model selection for stochastic block models
• Consistency and Asymptotic Normality of Stochastic Block Models Estimators from Sampled Data