Estimating And Testing In Partially Linear Panel Data Models With Fixed Effects

Panel data is the repeated observation data of the individual sections at different time points.Panel data has become a hot topic in the contemporary economic field.On the one hand,because its excellent characteristics,on the other hand,there are still many problems and unknown fields to explore in the application of panel data.In this paper,we consider the problem of estimating and testing in partially linear panel data model with fixed effects.Firstly,the difference method is used to eliminate the influence of fixed effects;Then we approximate the unknown parameter components in the model based on the B-spline function expansion method,the model is further transformed into a parametric model.and we use the least square method to gain the estimation of the parametric and nonparametric components,under some appropriate conditions,the estimator's asymptotic normality and convergence rate is proved;Finally,we construct appropriate statistic test to test serial correlation and heteroskedasticity problems.Numerical simulation studies analysis show that the estimation methods and test statistics presented perform well in this paper.This paper's arrangement is as follows.Chapter 1,the partially linear fixed effects panel data model(PLPDM)is proposed.The correlated knowledge of the PLPDM and the main work of this paper are introduced.Chapter 2,we give the B-spline estimation of the model and construct the test statistic of the serial correlation and heteroskedasticity.In Chapter 3,we give some hypothetical conditions and correlated theroms.Chapter 4,the finite sample characteristics of the proposed estimation methods and the constructed statistics are proved by numerical simulation.we give the detailed proof process of the theorems we proposed in this paper in chapter 5.