Modeling homophily and stochastic equivalence in symmetric relational data

Abstract

This article discusses a latent variable model for inference and prediction of symmetric relational data. The model, based on the idea of the eigenvalue decomposition, represents the relationship between two nodes as the weighted inner-product of node-specific vectors of latent characteristics. This ``eigenmodel'' generalizes other popular latent variable models, such as latent class and distance models: It is shown mathematically that any latent class or distance model has a representation as an eigenmodel, but not vice-versa. The practical implications of this are examined in the context of three real datasets, for which the eigenmodel has as good or better out-of-sample predictive performance than the other two models.

12 pages, 4 figures, 1 table

Cited by in corpus (17)

• The method of moments and degree distributions for network models
• Max-Margin Nonparametric Latent Feature Models for Link Prediction
• Bayesian latent structure discovery from multi-neuron recordings
• Dyadic data analysis with amen
• Sparse matrix-variate Gaussian process blockmodels for network modeling
• Network Cross-Validation for Determining the Number of Communities in Network Data
• Generalized linear models with low rank effects for network data
• An iterative step-function estimator for graphons
• Topics in social network analysis and network science
• Bayesian analysis of matrix data with rstiefel
• A Review of Latent Space Models for Social Networks
• Community detection in sparse latent space models
• Factor Models for High-Dimensional Dynamic Networks: with Application to International Trade Flow Time Series 1981-2015
• Network Mediation Analysis Using Model-based Eigenvalue Decomposition
• Multiplicative Coevolution Regression Models for Longitudinal Networks and Nodal Attributes
• A Latent Space Model for Multilayer Network Data
• An Eigenmodel for Dynamic Multilayer Networks