Linear Models And Empirical Analysis Of Panel Data

Panel Data are also called as TS/CS (Time Series/Cross Series). A panel consists of a sequence of observations, repeated through time, on a set of statistical units or "individuals" (persons, households, firms, regions, countries, etc.). It can reflect not only the law for a certain period each individual data also the change law over time to describe each individual. Set a time-series and cross-section data to the common advantage. Relative to the data analysis that only use time-series data or cross-section data, panel data model has many advantages: (1) the integration of variable cross-section and time variables can significantly reduce the problems caused by missing variables; (2) Panel-data model is usually a large amount of data available to researchers, this will increase the flexibility and reduce the collinearity among explanatory variables, thereby improve the effectiveness of estimates; (3) We can analyze of economic issues from Multi-level by panel data model. Because Panel data consist of time-series data and cross section data, therefore, its statistical properties are of features of time-series and cross-section. Therefore, analysis methods for Panel data are difficult that of time-series data or cross-sectional data.The dissertation consists of four chapters. In the first chapter, a general discussion on Panel Data analysis is given. Static panel data models are discussed in the second chapter, in this chapter, we studied estimation of parameters in static panel data models and test of models. In the third chapter, dynamic panel data models are discussed. In the forth chapter, the empirical analysis of Panel data are discussed, and a cause of panel data analysis is presented.