A0397
Title: Estimation of functional coefficient panel data models with endogenous selectivity and fixed effects
Authors: Alexandra Soberon - Universidad de Cantabria (Spain) [presenting]
Daniel Henderson - University of Alabama (United States)
Juan Manuel Rodriguez-Poo - Universidad de Cantabria (Spain)
Taining Wang - Capital University of Economics and Business (China)
Abstract: A novel estimation approach is developed for functional coefficient panel data models with sample selection and fixed effects. A two-step pairwise approach is proposed that avoids strict identification restrictions, and individual heterogeneity and selection bias are simultaneously addressed. The first stage estimates the selection equation parameters, while the second stage estimates the regression of interest using a generalized local weighting scheme that removes the sample selection bias asymptotically using the estimates of the previous stage. The asymptotic properties of the proposed estimators are established under rather weak assumptions, and the method's superior computational efficiency is demonstrated with respect to existing approaches and finite-sample performance through Monte Carlo simulations.
