Inference on nonparametric panel data models with fixed effects and censored dependent variables
提出了一种非参数删失面板回归模型的估计和推断方法，通过模拟和工资差距实证表明，该方法在结构函数非线性时优于参数模型。
.This article develops estimation and inference methods for a nonparametric generalization of a censored panel regression model with fixed effects. I demonstrate that the conventional trimmed LAD identification strategy continues to hold for identifying the nonparametric component of the generalized model. Building on the conditional moment condition implied by this strategy, I propose a plug-in Sieve Minimum Distance estimator for the model’s structural function. The proposed approach yields consistent and asymptotically normal estimators of the functionals of interest. Monte Carlo simulations assess the finite-sample performance of the estimation and inference procedures for the Average Partial Effect. The simulation results highlight the advantages of the proposed estimator and test statistics, particularly when the structural function is nonlinear and the proportion of censored observations is moderate. The empirical application revisits the top-coded wage equation used to study the evolution of the Black-White wage gap following Title VII of the Civil Rights Act of 1964 and Executive Order 11246. The new estimates suggest that changes in the wage gap were more modest than those implied by parametric model estimates.