Additive and Multiplicative Effects Network Models

Network datasets typically exhibit certain types of statistical patterns, such as within-dyad correlation, degree heterogeneity, and triadic patterns transitivity clustering. The first two these can be well represented with a social relations model, type additive effects model originally developed for continuous dyadic data. Higher-order multiplicative models, which are related to matrix decompositions that commonly used matrix-variate data analysis. Additionally, models generalize other popular latent feature network the stochastic blockmodel space model. In this article, we review general regression framework analysis combines effects, accommodates variety types, including continuous, binary ordinal relations.