Maximum lilkelihood estimation in the $\beta$-model

Abstract

We study maximum likelihood estimation for the statistical model for undirected random graphs, known as the $\beta$-model, in which the degree sequences are minimal sufficient statistics. We derive necessary and sufficient conditions, based on the polytope of degree sequences, for the existence of the maximum likelihood estimator (MLE) of the model parameters. We characterize in a combinatorial fashion sample points leading to a nonexistent MLE, and nonestimability of the probability parameters under a nonexistent MLE. We formulate conditions that guarantee that the MLE exists with probability tending to one as the number of nodes increases.

Published in at http://dx.doi.org/10.1214/12-AOS1078 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (6)

• Graph limits and exchangeable random graphs
• On the toric algebra of graphical models
• Maximum likelihood estimation in log-linear models
• A central limit theorem in the $\beta$-model for undirected random graphs with a diverging number of vertices
• Null models for network data
• Link and subgraph likelihoods in random undirected networks with fixed and partially fixed degree sequence

Cited by in corpus (22)

• Nonparametric graphon estimation
• Exponential-Family Models of Random Graphs: Inference in Finite-, Super-, and Infinite Population Scenarios
• On the Question of Effective Sample Size in Network Modeling: An Asymptotic Inquiry
• Asymptotics in directed exponential random graph models with an increasing bi-degree sequence
• Null models for network data
• Asymptotic normality in the maximum entropy models on graphs with an increasing number of parameters
• Concentration and consistency results for canonical and curved exponential-family models of random graphs
• Complex martingales and asymptotic enumeration
• Degree-based network models
• Consistent structure estimation of exponential-family random graph models with block structure
• The $\beta$-model for Random Graphs --- Regression, Cramér-Rao Bounds, and Hypothesis Testing
• Statistical Models for Degree Distributions of Networks
• Edge differentially private estimation in the $\beta$-model via jittering and method of moments
• Time-varying $\beta$-model for dynamic directed networks
• On the Geometry and Extremal Properties of the Edge-Degeneracy Model
• Signal Detection in Degree Corrected ERGMs
• Coauthorship and citation networks for statisticians: Comment
• An algebraic Monte-Carlo algorithm for the Partition Adjacency Matrix realization problem
• Monte Carlo goodness-of-fit tests for degree corrected and related stochastic blockmodels
• Toric algebra of hypergraphs
• Maximum likelihood geometry in the presence of data zeros
• Maximum likelihood degree of the $\beta$-stochastic blockmodel