Bayesian Inference on the Efficiency of Time-Varying Stochastic Frontier Model for Chinese Economy

• 1、School of Mathematics and Econometrics ,Hunan University,Changsha,Hunan 410081
• 2、School of Mathematics and Econometrics,Hunan University,Changsha,Hunan 410081

Abstract：Based on the time-varying stochastic production frontier model,this paper has studied the differences in production efficiency of various provinces and cities in China and time-varying effects of production factors on economic output elasticity,and conducted an empirical research on the annual data of 31 provinces and cities from 2010 to 2017 using Bayesian statistical method. The study found that the production efficiency shows a downward trend overall, and the production efficiency has some differences between regions.With the passage of time, the output of capital to economy in the western region decreased gradually, while the labor force increased gradually. Per capita education years promotes the economic output of the eastern region, but its effectiveness gradually weakens. Environmental pollution has a significant negative effect on the economic production in the central and eastern regions, and its influence on the eastern region gradually weakens.

Keywords： Bayesian statistical Time-varying stochastic frontier model Production efficiency analysis Gibbs sampling