Problem
Research questions and friction points this paper is trying to address.
Bipartite Networks
Dynamic Networks
Longitudinal Data
Network Modeling
Latent Structure
Modeling Bipartite Dynamic Networks: An Additive and Multiplicative Effects Model
🤖 AI Summary
This study addresses the challenge of modeling dependence structures in dynamic bipartite networks and the limitations of traditional models in capturing latent multiplicative effects within longitudinal interactions. We propose an Additive and Multiplicative Effects (AME) framework for longitudinal bipartite data. Methodologically, we introduce a novel dynamic bipartite AME model that disentangles latent multiplicative structures from observed covariates. To enhance computational efficiency, we integrate block coordinate descent with the squared iterative method and implement the proposed approach in the R package RAMEN. Simulation studies and an empirical application to global production networks demonstrate that the model significantly improves coefficient estimation accuracy and successfully reveals evolutionary patterns of country participation that static models fail to capture.
📝 Abstract
Researchers frequently study interactions between two distinct types of actors, represented as bipartite networks. These networks exhibit dependence patterns that differ from those in one-mode networks and therefore require models tailored to their structure. This paper develops an additive and multiplicative effects (AME) framework for longitudinal bipartite data. First, I distinguish the dependence structure and specify the corresponding modeling assumptions. Second, I introduce the bipartite dynamic AME model and develop an estimation procedure based on block coordinate descent. Third, I incorporate a squared iterative method to improve computational efficiency. Using simulations and an application to global production networks, I show that the model improves coefficient estimation, more accurately recovers the data-generating process, and better captures the multiplicative latent structure. The model reveals evolving patterns in countries' global production engagement that are not captured by observed covariates or static specifications. I provide an R package, RAMEN, to facilitate implementation.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Bipartite dynamic networks
Additive and multiplicative effects model
Block coordinate descent
Squared iterative method
Latent structure
🔎 Similar Papers
No similar papers found.
