Random Coefficients on Endogenous Variables in Simultaneous Equations Models

摘要: This article considers a classical linear simultaneous equations model with random coefficients on the endogenous variables. Simultaneous models are used to study social interactions, strategic interactions between firms, and market equilibrium. Random coefficient allow for heterogeneous marginal effects. I show that seemingly unrelated regression common regressors not point identified, which implies identified. Important features of these models, however, can be For two-equation systems, give two sets sufficient conditions identification coefficients’ distributions conditional exogenous covariates. The first allows small support continuous instruments under tail restrictions unobservables necessary identification. second requires full instruments, but nearly arbitrary unobservables. discuss how generalize results many equation where focus linear-in-means interaction propose consistent nonparametric kernel estimator based arguments. apply my Add Health data analyse peer effects in education.