Statistical Inference for Multilayer Networks in Political Science

Abstract

<p>Interactions between units in political systems often occur across multiple relational contexts. These relational systems feature interdependencies that result in inferential shortcomings and poorly-fitting models when ignored. General advancements in inferential network analysis have improved our ability to understand relational systems featuring interdependence, but developments specific to working with interdependence that cross relational contexts remain sparse. In this paper, I introduce a multilayer network approach to modeling systems comprising multiple relations using the exponential random graph model. In two substantive applications, the first a policy communication network and the second a global conflict network, I demonstrate that the multilayer approach affords inferential leverage and produces models that better fit observed data.</p>

Informal Diplomacy Reinforces Formal International Cooperation: Evidence from Track Two Diplomacy

“…Network approaches are useful for studies in international relations (Cranmer and Desmarais 2016), and prior work have found network effects to play important roles in international diplomacy (Kinne 2014;Maliniak and Plouffe 2011). I advance the use of networks in diplomatic studies by using recent developments in network science to jointly model formal and informal diplomacy as a multilayer network (Chen 2019).…”

“…These factors include covariates at the node and dyad level that are used in the classical regression framework, but also encompass network effects that capture the tendency for tie formation to depend on the presence of other ties in the network. 3 Traditionally, ERGMs are limited by the assumption that all ties on the network are the same, but recent developments have extended the ERGM to multilayer networks (Chen 2019), which allows heterogeneous relational data to be jointly modelled. This extension to ERGMs is particularly suitable for my study, as the mutual reinforcement hypothesized to exist between different layers of the joint track one and track two multiplex network is a type of network effect, which I describe in more detail below.…”

“…Besides being one of the first statistical examinations of this topic, my study makes a number of additional contributions to our understanding of track two effectiveness, and more generally how informal diplomacy facilitates international cooperation. First, I take advantage of recent developments in multilayer network modeling that is particularly suitable for analyzing the relationship between diplomatic cooperation across different tracks and temporal contexts (Chen 2019). Next, whereas the majority of prior track two research focused on the Middle East, I look at the Asia Pacific region.…”

Peer Review #2 of "A survey on exponential random graph models: an application perspective (v0.1)"

“…(Pan, 2018) studied these networks to answer the underlying questions about them and their effects on related subjects. Even further, some works such as (Chen, 2018) took the use of ERGM networks in modeling political networks a step further by incorporating multilayer networks properties into their models. He proved with experimental results that this multilayer approach toward ERGMs could better fit the model to the observed data.…”