Font size: Small Medium Large | Share to:


A Partial Identification Subnetwork Approach to Discrete Games in Large Networks: An Application to Quantifying Peer Effects

  • Date 2017-06-20 (Tue)
  • Time 02:30 PM
  • Venue Conference Room B110
  • Presider Professor Yu-Chin Hsu
  • Speaker Professor Tong Li
  • Background Professor Li received his Ph.D. from the University of Southern California in 1997. He is currently the Gertrude Conaway Vanderbilt Professor of Economics of Vanderbilt University. Professor Li’s primary research and teaching interests are microeconometrics with a focus on identification and inference of econometric models with latent variables, and game-theoretic models. He also studies dynamic/nonlinear panel data analysis, and empirical microeconomics with a focus on empirical analysis of strategic behavior of agents with asymmetric information.
  • Abstract This paper studies identification and estimation of discrete games in large networks, with an application to peer effects on smoking in friend networks. Due to the presence of multiple equilibria, the model is not point identified. We adopt the partial identification approach by constructing moment inequalities on choice probabilities of subnetworks. Doing so not only significantly reduces the computational cost, but also enables us to find consistent estimator of the moment conditions even when the network is large and the friendship relationship structure varies significantly among networks. Monte Carlo studies are conducted to evaluate the performance of the subnetwork approach. In the application using the Add Health data, we find significant and positive peer effects on smoking.