应用经验生产函数测算时间序列决策单元有效性的方法

Data Envelopment Analysis Method Based on Time Series Data

数据包络分析方法构建的基础是应用决策单元的指标数据来模拟经验生产函数,而生产函数本身又是描绘在技术水平不变情况下各种生产要素与所能生产的最大产出之间的关系。对于一组时间序列数据而言,在不同时间点上决策单元的技术水平一般会发生变化,所以DEA方法是否可以测算时间序列决策单元效率一直存在质疑和分歧。为了寻找DEA方法测算时间序列决策单元效率的理论基础和更有效的测算方法,本文首先给出了时间序列DEA模型成立的条件和基础。然后,从经验生产函数的构造出发,给出了三种测算时间序列决策单元的DEA模型,并与目前存在的各种时间序列DEA模型进行了比较研究,从理论上讨论了不同时间序列DEA模型给出的效率值之间的关系。同时,改进了测算时间序列决策单元技术进步的方法。最后,应用本文结论分析了广东省1985—2013年的经济发展效率问题。

The basis of the data envelopment analysis(DEA)method is using the data of decision making units(DMUs)to simulate the empirical production function,and the production function is used to describe the relationship between various factors of production and its maximum output under constant technical level.For a set of time series data,it is often questioned whether the DEA method can be used to calculate the efficiency of time series DMU because the technical level of decision making unit is generally changed at different time points.In order to find the theoretical basis and more effective calculation method that DEA method is used to calculate the efficiency of time series DMU,we first give the condition and basis for establishment of DEA model.Then,based on the construction of empirical production function,we give three new DEA models to calculate the productivity of decision-making unit by using time series data.At the same time,we also make a comparative study of various time series DEA models,and theoretically discuss the relationship between the efficiency values given by different time series DEA models.In addition,the method of measuring the technical progress is improved.Finally,the efficiency of economic development in Guangdong province from 1985 to 2013 is analyzed.