Social Interactions: A Game Theoretic Approach

This paper uses a game theoretic model to capture the interactions among individuals within a social network, and establishes nonparametric identification and inference on the game structural model. Consider observations from a single equilibrium of a network game in which each player chooses an action from a finite set and is subject to interactions that are local --- the interactions only occur among friends. All observations are potentially dependent on each other because they are interpreted as arising from a single equilibrium of settings where players interact directly or indirectly. Simple assumptions about the structure are made that ensure that the game has a unique equilibrium and the equilibrium has a stability property. The formulation of this stability property is new and serves as the basis for statistical inference. I establish the identification of the structural model and introduce an estimation procedure called (sieve) maximum approximated likelihood.