New model tracks changing interactions between two actor groups over time
Modeling Bipartite Dynamic Networks: An Additive and Multiplicative Effects Model
Social and Information Networks
Summary
Some studies look at how two different groups, like countries and products, connect over time. The paper creates a new way to model these changing connections by combining simpler additive parts and hidden patterns that multiply together. The authors also made the method faster to compute. Their model better understands how the relationships evolve and finds patterns that other models miss. They shared their approach as an easy-to-use software package for others.
What this means in practice
- •For economic analysts: Analyze how countries' trade relationships evolve over time with products using a detailed dynamic network model.
- •For supply chain managers: Improve understanding of changing supplier and product dynamics by applying advanced bipartite network models to supply data.
Authors
Jing Luo
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.