Social Interactions Quantiles

This paper provides a method to study heterogenous social interactions in a large network. The method is based on a maximum score model with social interactions where the social interactions are modeled in a simultaneous discrete game. The heterogenous social interactions parameters are estimated by a nested pseudo score (NPS) approach which is developed to tackle the computational burden pertaining to a discrete game with a large number of players. Uniform consistency and asymptotic normality are established for the proposed NPS estimator. I apply the method to study the heterogenous peer effects among students on exercise decisions, using the National Longitudinal Study of Adolescent Health dataset