Research And Application Of Semiparametric Time-varying Coefficient Stochastic Frontier Models

The measurement of technical efficiency is a very important issue in the field of economics and management.As one of the main methods of measuring technical efficiency,the stochastic frontier model has received in-depth theoretical research and extensive practical applications.In the existing research of stochastic frontier analysis,the relationship between the dependent variable(output)and the independent variable(input)in the production function assumes a logarithmic linear form,which is often not satisfied in actual problem research.The incorrect setting of regression function will lead to the subsequent inferred result deviating from the actual situation and obtaining the conclusion inconsistent with the actual situation.In order to better explore the real relationship between dependent variables and independent variables and improve the flexibility and adaptability of the model,several types of nonparametric and semi-parametric stochastic frontier models have been proposed and studied in recent years.In this paper,a new class of semi-parametric time-varying coefficient panel data model is proposed based on the existing semi-parametric stochastic frontier model and its statistical inference and application are studied.In terms of statistical inference of the model,this paper uses kernel estimation and local linear methods to estimate the unknown coefficient function in the model.In order to test whether individual effects and regression coefficients change significantly over time,the paper constructs a generalized Ftest statistic.In addition,the paper also discusses a generalized form of the proposed model using the profile least square method to estimate.In the application of the model,Based on the panel observation data of 29 provinces in my country from 2000 to 2017,this paper uses the proposed model to study the regional carbon emission efficiency in China,and reached meaningful conclusions.With the help of semi-parametric modeling method,the paper proposes a new type of semi-parametric panel stochastic frontier model.The research results enrich the research content of semi-parametric model and stochastic frontier analysis,and provide a scientific statistical analysis method for the measurement of technical efficiency.