部分线性单指数动态面板模型的惩罚二次推断函数估计

本文将个体效应与模型的线性部分和非线性部分同时进行估计，构建了惩罚二次推断函数估计法，估计了个体内具有自相关结构的随机效应部分线性单指数动态面板模型。该方法的优点是能够克服由个体效应引起的内生性问题，因此无论滞后因变量处于线性部分还是处于非参数部分均可得到一致有效的估计。进一步，本文证明了该估计方法的一致性和渐近正态性，同时还用Monte Carlo模拟实验评估了该估计方法在有限样本下的表现。

Penalized Quadratic Inference Functions Estimation for Partially Linear Single Index Dynamic Panel Model

This paper proposes penalized quadratic inference function estimation method for partially linear single index dynamic panel model with random effects which has autocorrelation structure within individuals. The method estimates individual effects and parametric and nonparametric parts of model together. Its advantage is that it can overcome endogenous which is caused by induvitual effects. So the estimation method in this paper can be applicable to the case which lagged dependent variable either in linear part or non-parametric part. Moreover,this paper derives consistency and asymptotic normality of the proposed method in this paper and evaluates the finite sample performance of the proposed method via Monte Carlo simulation.