Cross-sectional Dependence Test Under Serial Correlation In Panel Data Models

Recently,the cross-sectional dependence test is one of the hot topics in panel data model analysis.The traditional CD_lmtest suppose that the disturbance terms does not have serial correlation.However,serial correlation is a common problem in panel data analysis.Therefore,we modify the traditional CD_lmtest so that the adjusted test statistics can effectively test the cross-sectional dependence under arbitrary serial correlation structure.Under appropriate conditions,we establish the theoretical properties of the proposed adjusted CD_lmtest statistic is asymptotically normal.Simulation studies show that the finite sample performance of the test statistic performs well with respect to the size and the power.This paper's arrangement is as follows.Chapter 1,the fixed effects panel data model is proposed.The correlated knowledge of the cross-sectional dependence and the main work of this paper are introduced.Chapter 2 introduces the model structure and modify the traditional CD_lmtest to construct a new test statistic.In Chapter 3,we give some hypothetical conditions and correlated theorems.Simulation studies are presented in Chapter4 to illustrate an the finite sample performance of the proposed test.Finally,we give the detailed proof process of the theorems we proposed in this paper in chapter 5.