Adam Smith Business School

Bayesian Inference of Network Formation Models with Payoff Externalities

Friday, 28th October, 3 pm-4 pm

Santiago Montoya Blandon is a Lecturer (Assistant Professor) in Econometrics for the Adam Smith Business School at the University of Glasgow. His research covers theoretical and applied econometrics, focusing on Bayesian analysis, and attempts to present rigorous solutions to problems that arise in applied microeconomic settings. He is particularly interested in the econometric challenges inherent to network and spatial data.

Abstract

This paper provides a novel approach to identifying and performing inference on the utility parameters of a network formation model with payoff externalities using observed network data. The existence of externalities induces an issue of multiple equilibria. We first show that local point identification of the parameters of interest is possible under a mild assumption on the equilibrium selection process. We propose a Bayesian estimation method to conduct statistical inference of structural payoff coefficients. Our methods also allow us to estimate equilibrium selection probabilities, but this requires repeated networks. We address this issue by proposing a composite likelihood function based on subgraphs that compose the observed network. We show that using a composite likelihood induces misspecification and characterises the Kullback-Leibler divergence that measures this misspecification error. We present an empirical application to model the network formation process of individuals creating social connections in villages in Karnataka, India and find strong evidence of homophily effects.

 

---

For further information, please get in touch with business-school-research@glasgow.ac.uk  

First published: 11 October 2022