Mixed Logit Models and Network Formation

TL;DR

This paper introduces the repeated-choice (RC) model as a superior method for studying network formation, demonstrating its advantages over the traditional multinomial logit (MNL) model through synthetic and real-world network analyses.

Contribution

The paper proposes the RC model for network formation, showing it overcomes limitations of MNL and provides more accurate and insightful analysis of network data.

Findings

RC model estimates synthetic networks more accurately than MNL.

In patent citation networks, RC reveals that new patents cite older, highly cited, and similar patents.

RC model offers valuable insights into sequential network formation processes.

Abstract

The study of network formation is pervasive in economics, sociology, and many other fields. In this paper, we model network formation as a `choice' that is made by nodes in a network to connect to other nodes. We study these `choices' using discrete-choice models, in which an agent chooses between two or more discrete alternatives. We employ the `repeated-choice' (RC) model to study network formation. We argue that the RC model overcomes important limitations of the multinomial logit (MNL) model, which gives one framework for studying network formation, and that it is well-suited to study network formation. We also illustrate how to use the RC model to accurately study network formation using both synthetic and real-world networks. Using edge-independent synthetic networks, we also compare the performance of the MNL model and the RC model. We find that the RC model estimates the data-generation process of our synthetic networks more accurately than the MNL model. In a patent citation network, which forms sequentially, we present a case study of a qualitatively interesting scenario -- the fact that new patents are more likely to cite older, more cited, and similar patents -- for which employing the RC model yields interesting insights.