Variable Selection Of Two Kinds Of Spatial Panel Data Models

In recent decades,with the continuous progress of science and technology,a large number of spatial panel data generated by people's life or economic behavior have been collected and recorded.Spatial panel data model is one of the important models for mining spatial panel data information.When building spatial panel data model,it has become a research hotspot of econometrics to select key variables in an appropriate way to make spatial panel data model have better explanatory ability.In this context,this paper combines the spatial panel data model with the theoretical knowledge of variable selection to study the variable selection of linear spatial panel autoregressive model and spatial panel error model.The main content of this paper is divided into four chapters:The first chapter is the introduction,which briefly introduces the research background,spatial panel data model and variable selection method,as well as the main content and structure of this paper.The second chapter introduces the definition of spatial panel data model,several definitions of spatial weight matrix,and several common variable selection methods.In the third chapter,we study the adaptive lasso variable selection problem of first-order spatial panel autoregressive model,and prove that the method has good asymptotic properties under certain conditions.The finite sample performance of the method is verified by Monte Carlo method.In Chapter 4,the adaptive lasso variable selection of the first-order spatial panel error model is studied.The results of Monte Carlo simulation show that the method can effectively identify zero and non-zero coefficients.