Consistency under sampling of exponential random graph models

DOI10.1214/12-AOS1044zbMATH Open1269.91066arXiv1111.3054OpenAlexW3103377809WikidataQ40745745 ScholiaQ40745745MaRDI QIDQ355086FDOQ355086

Authors: Cosma Rohilla Shalizi, Alessandro Rinaldo 

Publication date: 24 July 2013

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: The growing availability of network data and of scientific interest in distributed systems has led to the rapid development of statistical models of network structure. Typically, however, these are models for the entire network, while the data consists only of a sampled sub-network. Parameters for the whole network, which is what is of interest, are estimated by applying the model to the sub-network. This assumes that the model is consistent under sampling, or, in terms of the theory of stochastic processes, that it defines a projective family. Focusing on the popular class of exponential random graph models (ERGMs), we show that this apparently trivial condition is in fact violated by many popular and scientifically appealing models, and that satisfying it drastically limits ERGM's expressive power. These results are actually special cases of more general results about exponential families of dependent random variables, which we also prove. Using such results, we offer easily checked conditions for the consistency of maximum likelihood estimation in ERGMs, and discuss some possible constructive responses.

Full work available at URL: https://arxiv.org/abs/1111.3054

Recommendations

• Estimating and understanding exponential random graph models
• On the geometry of discrete exponential families with application to exponential random graph models
• Conditional marginalization for exponential random graph models
• Consistent structure estimation of exponential-family random graph models with block structure
• Perspectives on exponential random graphs

zbMATH Keywords

exponential familynetwork modelsnetwork samplingindependent incrementsexponential random graph modelprojective familysufficient statistics

Mathematics Subject Classification ID

Processes with independent increments; Lévy processes (60G51) Inference from stochastic processes (62M99) Non-Markovian processes: estimation (62M09) Applications of statistics to social sciences (62P25) Social networks; opinion dynamics (91D30) Sufficient statistics and fields (62B05)

Cites Work

• A nonparametric view of network models and Newman–Girvan and other modularities
• Graphical models, exponential families, and variational inference
• Limits of dense graph sequences
• Title not available (Why is that?)
• A survey of statistical network models
• Statistical analysis of network data. Methods and models
• Title not available (Why is that?)
• An Exponential Family of Probability Distributions for Directed Graphs
• Adjusting for network size and composition effects in exponential-family random graph models
• Logit models and logistic regressions for social networks. I: An introduction to Markov graphs and \(p^*\)
• Modeling social networks from sampled data
• On the geometry of discrete exponential families with application to exponential random graph models
• Discrete temporal models of social networks
• Markov Graphs
• Title not available (Why is that?)
• Networks, crowds and markets. Reasoning about a highly connected world.
• Networks. An introduction.
• Graph limits and exchangeable random graphs
• Estimating and understanding exponential random graph models
• Title not available (Why is that?)
• Consistency under sampling of exponential random graph models
• Maximum likelihood estimation in the \(\beta\)-model
• Random graphs with a given degree sequence
• Large deviations
• Theory of statistics
• Title not available (Why is that?)
• The Role of Sufficiency and of Estimation in Thermodynamics
• Extremal families and systems of sufficient statistics
• Title not available (Why is that?)
• Kepler's area law in the \textit{Principia}: filling in some details in Newton's proof of Proposition 1.
• Projective limit random probabilities on Polish spaces
• Exchangeable Rasch matrices
• A candidate's formula: A curious result in Bayesian prediction
• Title not available (Why is that?)
• On the bias of traceroute sampling
• Renormalization group and probability theory
• Applications of Stein's method for concentration inequalities
• Mixing time of exponential random graphs
• Maximum entropy Gaussian approximations for the number of integer points and volumes of polytopes

Cited In (54)

• Longitudinal network models and permutation‐uniform Markov chains
• Networks beyond pairwise interactions: structure and dynamics
• On the question of effective sample size in network modeling: an asymptotic inquiry
• Consistency under sampling of exponential random graph models
• Affiliation discrete weighted networks with an increasing degree sequence
• Signal detection in degree corrected ERGMs
• Sub-critical exponential random graphs: concentration of measure and some applications
• Variable Selection for High-Dimensional Nodal Attributes in Social Networks with Degree Heterogeneity
• Crawling subsampling for multivariate spatial autoregression model in large-scale networks
• Statistics of the two star ERGM
• Multiresolution Network Models
• On the geometry of discrete exponential families with application to exponential random graph models
• Degeneracy in sparse ERGMs with functions of degrees as sufficient statistics
• Exponential-family models of random graphs: inference in finite, super and infinite population scenarios
• Asymptotic structure of constrained exponential random graph models
• Outliers and influential observations in exponential random graph models
• Modelling and monitoring social network change based on exponential random graph models
• Graphical construction of spatial Gibbs random graphs
• Time-varying network models
• Approximate Bayesian computation for exponential random graph models for large social networks
• A motif building process for simulating random networks
• Adjustable network reconstruction with applications to CDS exposures
• Bootstrapping exchangeable random graphs
• Sparse maximum-entropy random graphs with a given power-law degree distribution
• A statistical framework for modern network science
• Concentration and consistency results for canonical and curved exponential-family models of random graphs
• Consistent structure estimation of exponential-family random graph models with block structure
• Link Prediction for Egocentrically Sampled Networks
• Projective, sparse and learnable latent position network models
• Projective families of distributions revisited
• Limits of multi-relational graphs
• Conditional estimation of exponential random graph models from snowball sampling designs
• A simple and efficient algorithm for modeling modular complex networks
• Minimum distance estimators of population size from snowball samples using conditional estimation and scaling of exponential random graph models
• Estimating and understanding exponential random graph models
• An introduction to large deviations for random graphs
• Exponential random simplicial complexes
• Asymptotic normality in the maximum entropy models on graphs with an increasing number of parameters
• Consistency of maximum likelihood for continuous-space network models. I
• A perfect sampling method for exponential family random graph models
• Inference in a probit model for affiliation networks
• Asymptotics in directed exponential random graph models with an increasing bi-degree sequence
• Asymptotic distribution in affiliation finite discrete weighted networks with an increasing degree sequence
• On the time to identify the nodes in a random graph
• Adjusting for network size and composition effects in exponential-family random graph models
• Practical Network Modeling via Tapered Exponential-Family Random Graph Models
• Statistical Inference in a Directed Network Model With Covariates
• Conditional marginalization for exponential random graph models
• Bayesian Shrinkage for Functional Network Models, With Applications to Longitudinal Item Response Data
• Random Partition Models for Microclustering Tasks
• Edge Exchangeable Models for Interaction Networks
• Social network mediation analysis: a latent space approach
• Estimating contact network properties by integrating multiple data sources associated with infectious diseases
• Investigation of patient-sharing networks using a Bayesian network model selection approach for congruence class models

Uses Software

• statnet

This page was built for publication: Consistency under sampling of exponential random graph models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q355086)