Identifying distributional characteristics in random coefficients panel data models

We study the identification of panel models with linear individual-specific coefficients when T is fixed. We show identification of the variance of the effects under conditional uncorrelatedness. Identification requires restricted dependence of errors, reflecting a trade-off between heterogeneity and error dynamics. We show identification of the probability distribution of individual effects when errors follow an Autoregressive Moving Average process under conditional independence. We discuss Generalized Method of Moments estimation of moments of effects and errors and construct non-parametric estimators of their densities. As an application, we estimate the effect that a mother smoking during pregnancy has on her child's birth weight.