摘要
DEA-Malmquist方法能在样本数量不大时对全要素生产率(TFP)的增长进行分解。然而中国各省份间要素禀赋各异、经济发展水平差距较大,不能满足其"所有经济体面对同一技术前沿"的假设。通过估算DEA-Malmquist方法与索洛余值方法的偏差发现,该方法适用于中国省份TFP的估计。利用省份面板数据进行的实证研究表明,对DEA-Malmquist方法估计的省份结果取几何平均,可以得到与传统方法估计的全国TFP增长一致的结果;与传统方法相比,分解TFP得到的技术效率变化曲线更符合中国经济制度的变化历程。
DEA-Malmquist is able to decompose the Total Factor Productivity （TFP） when the sample size is not large. However, since the factor endowment condition and developing level varies a lot among the provinces of China, it seems hard to meet the assumption of ＂all economies face the same technology frontier＂. We estimate the difference between the results of DEA-Malmquist method and Solow Residual method. Furthermore, we propose to use the geometric re＇can of provinces results to represent the TFP estimation and decomposition of China. Empirical study show that our method coordinates with traditional methods. Moreover, the estimated technology efficiency better fits the change of China＇ s economic system. The technology efficiency, one of the factors decomposed from TFP, is consistent with the variation of the potential output efficiency.
出处
《贵州财经学院学报》
北大核心
2013年第1期32-39,共8页
Journal of Guizhou College of Finance and Economics
