The Quantile Regression Method Research Of C-D Production Function Parameter Estimation

At present the research of production function model is quite complete. This model is mainly applied for empirical analysis of input factors in the economic growth. The content is becoming more deepened and enrich, which included institutional factors, industry structure factors, human resource factors. Least square method is the major estimation method about production function model parameter. The quantile regression is used less and just stay in the level of empirical analysis. In this thesis, the Cobb-Douglas production function (CD production function) model with two explanatory variables is used as an example to do quantile regression method research of cross-sectional data and panel data production function model parameters estimated. On the basis of the results of the analysis of scholars, quantile regression method of cross-sectional data and panel data production function model is proposed respectively through theoretical methods and Monte Carlo simulation. A comparative method is also applied with the classical least squares regression method. The purpose of this thesis is to supply methods for parameter estimation problems of production function model. References are also provided for parameter estimation process of C-D production function.Meanwhile an empirical study has been researched of production function model based on cross-sectional data of Beijing in2007and panel data of Chinese mainlan provinces from1978to2008. The overal distribution characteristics of the economic growth of Beijing and China can be reflected through the quantile regression method, which can be used by the government for understanding the overall growth and appropriate decision.The meaning of the research is listed as following. Firstly, quantile regression method makes up the shortage of least square method about estimation parameter of production function model. Next, different angles, such as cross-sectional data panel data quantile regression can be used for quantile regression research. Last but not the least, empirical analysis based on the Beijing cross-sectional data and panel data analysis reflecting the various elements of Beijing and Chinese economic growth, and relevant conclusions and policy recommendations through quantile regression.