Separating the wheat from the chaff: Bayesian regularization in dynamic social networks

In recent years there has been an increasing interest in the use of relational event models for dynamic social network analysis. The basis of these models is the concept of an event, defined as a triplet of time, sender, and receiver of some social interaction. The key question that relational event models aim to answer is what drives the pattern of social interactions among actors. Researchers often consider a very large number of predictors in their studies (including exogenous effects, endogenous network effects, and interaction effects). However, employing an excessive number of effects may lead to overfitting and inflated Type-I error rates. Moreover, the fitted model can easily become overly complex and the implied social interaction behavior difficult to interpret. A potential solution to this problem is to apply Bayesian regularization using shrinkage priors to recognize which effects are truly nonzero (the wheat) and which effects can be considered as (largely) irrelevant (the chaff). In this paper, we propose Bayesian regularization methods for relational event models using four different priors for both an actor and a dyad relational event model: a flat prior model with no shrinkage, a ridge estimator with a normal prior, a Bayesian lasso with a Laplace prior, and a horseshoe prior. We apply these regularization methods in three different empirical applications. The results reveal that Bayesian regularization can be used to separate the wheat from the chaff in models with a large number of effects by yielding considerably fewer significant effects, resulting in a more parsimonious description of the social interaction behavior between actors in dynamic social networks, without sacrificing predictive performance.

Automated summary

In recent years, relational event models have gained increasing interest for dynamic social network analysis. These models are based on the concept of an event, which is defined as a triplet of time, sender, and receiver of a social interaction. The goal of relational event models is to understand the factors driving the pattern of social interactions among actors. Researchers often include a large number of predictors in their studies, but this can lead to overfitting and complex models that are difficult to interpret. Bayesian regularization methods offer a potential solution by identifying the relevant effects and reducing the number of significant effects in the model. This paper proposes Bayesian regularization methods for relational event models and applies them to three empirical applications, showing that these methods can provide a more parsimonious description of social interaction behavior without sacrificing predictive performance.