Peer Effects in Binary Outcomes: Strategic Complementarity and Taste for Conformity With Endogenous Networks
构建了一个基于网络博弈的二元结果同伴效应模型，同时考虑策略互补性和从众偏好，推导了理性预期下贝叶斯纳什均衡的唯一性条件，并提出了非参数估计方法，应用于印度小额信贷参与研究。
ABSTRACT This paper introduces a generalized model of peer effects for binary outcomes, based on a network game that accounts for strategic complementarity (influence of the number of peers that select the same action) and conformity to social norms (penalizing deviations from the average peers' action). We derive conditions for a unique Bayesian Nash equilibrium under rational expectations about peers' behavior. Unobserved individual heterogeneity affecting network formation and peer effects is addressed nonparametrically using a control function nested pseudo likelihood estimator, whose asymptotic properties are established. The model is applied to microfinance participation in India.