Exponential-Family Models of Random Graphs: Inference in Finite-, Super-, and Infinite Population Scenarios

Abstract

Exponential-family Random Graph Models (ERGMs) constitute a large statistical framework for modeling sparse and dense random graphs, short- and long-tailed degree distributions, covariates, and a wide range of complex dependencies. Special cases of ERGMs are generalized linear models (GLMs), Bernoulli random graphs, $\beta$-models, $p_1$-models, and models related to Markov random fields in spatial statistics and other areas of statistics. While widely used in practice, questions have been raised about the theoretical properties of ERGMs. These include concerns that some ERGMs are near-degenerate and that many ERGMs are non-projective. To address them, careful attention must be paid to model specifications and their underlying assumptions, and in which inferential settings models are employed. As we discuss, near-degeneracy can affect simplistic ERGMs lacking structure, but well-posed ERGMs with additional structure can be well-behaved. Likewise, lack of projectivity can affect non-likelihood-based inference, but likelihood-based inference does not require projectivity. Here, we review well-posed ERGMs along with likelihood-based inference. We first clarify the core statistical notions of "sample" and "population" in the ERGM framework, and separate the process that generates the population graph from the observation process. We then review likelihood-based inference in finite-, super-, and infinite-population scenarios. We conclude with consistency results, and an application to human brain networks

References in corpus (51)

• Emergence of scaling in random networks
• Spectral clustering and the high-dimensional stochastic blockmodel
• High-dimensional Ising model selection using ${\ell }_1$-regularized logistic regression
• A Separable Model for Dynamic Networks
• Modeling social networks from sampled data
• Consistency of spectral clustering in stochastic block models
• Fast community detection by SCORE
• Exponential-Family Random Graph Models for Valued Networks
• Consistency of community detection in networks under degree-corrected stochastic block models
• Pseudo-likelihood methods for community detection in large sparse networks
• Estimating and understanding exponential random graph models
• Stochastic blockmodels with growing number of classes
• Modeling homophily and stochastic equivalence in symmetric relational data
• Rate-optimal graphon estimation
• Latent Space Models for Dynamic Networks
• Sparse graphs using exchangeable random measures
• Random graphs with a given degree sequence
• Solution of the 2-star model of a network
• The method of moments and degree distributions for network models
• Exponential Random Graph Modeling for Complex Brain Networks
• Statistical Inference for Valued-Edge Networks: Generalized Exponential Random Graph Models
• Covariate-assisted spectral clustering
• Analyzing complex functional brain networks: fusing statistics and network science to understand the brain
• Consistency under sampling of exponential random graph models
• Solution for the properties of a clustered network
• Adjusting for Network Size and Composition Effects in Exponential-Family Random Graph Models
• On Russian Roulette Estimates for Bayesian Inference with Doubly-Intractable Likelihoods
• Maximum lilkelihood estimation in the $\beta$-model
• Asymptotics in directed exponential random graph models with an increasing bi-degree sequence
• On the Question of Effective Sample Size in Network Modeling: An Asymptotic Inquiry
• The Class of Random Graphs Arising from Exchangeable Random Measures
• Inference using noisy degrees: Differentially private $\beta$-model and synthetic graphs
• Universally consistent vertex classification for latent positions graphs
• Phase transitions in exponential random graphs
• Asymptotic normality in the maximum entropy models on graphs with an increasing number of parameters
• Edge-exchangeable graphs and sparsity
• Concentration and consistency results for canonical and curved exponential-family models of random graphs
• Exponential-Family Random Graph Models for Rank-Order Relational Data
• A statistical model for brain networks inferred from large-scale electrophysiological signals
• Sampling perspectives on sparse exchangeable graphs
• On edge exchangeable random graphs
• A Dynamic Process Interpretation of the Sparse ERGM Reference Model
• Consistent structure estimation of exponential-family random graph models with block structure
• A Perfect Sampling Method for Exponential Family Random Graph Models
• Asymptotic quantization of exponential random graphs
• On the asymptotics of constrained exponential random graphs
• Weighted Exponential Random graph models: Scope and large network limits
• Estimating Subgraph Frequencies with or without Attributes from Egocentrically Sampled Data
• Bayesian Inference from Non-Ignorable Network Sampling Designs
• Phase transition in the two star Exponential Random Graph Model
• DERGMs: Degeneracy-restricted exponential random graph models

Cited by in corpus (10)

• Exponential random graph model parameter estimation for very large directed networks
• Consistent structure estimation of exponential-family random graph models with block structure
• Large-scale estimation of random graph models with local dependence
• "Predicting" after peeking into the future: Correcting a fundamental flaw in the SAOM -- TERGM comparison of Leifeld and Cranmer (2019)
• A Tale of Two Datasets: Representativeness and Generalisability of Inference for Samples of Networks
• Testing biological network motif significance with exponential random graph models
• Online network monitoring
• Highly Scalable Maximum Likelihood and Conjugate Bayesian Inference for ERGMs on Graph Sets with Equivalent Vertices
• Continuous Time Graph Processes with Known ERGM Equilibria: Contextual Review, Extensions, and Synthesis
• Likelihood-based Inference for Exponential-Family Random Graph Models via Linear Programming