三阶段组合效率测度模型与技术研发效率测度

发布时间：2018-07-26 15:33

【摘要】：在系统梳理现有可调整环境因素或/与统计噪音影响的三阶段组合效率测度模型的基础上,利用幅度调整测度(RAM)与随机前沿分析(SFA)结合,构建了无需强制性调整环境因素与统计噪音影响的三阶段组合效率测度改进模型——RAM-SFA-RAM.相对现有模型,新的组合测度模型不但在估计影响与调整投入和产出时可实现较为客观地设定环境因素与松弛变量之间存在非线性生产函数关系,以降低效率估计偏误,而且充分利用了RAM模型全面测度无效的优点,为从投入和产出两个方面同时过滤环境因素与统计噪音的差异性影响提供了新的分析途径;更重要的是,选用的RAM具有平移不变性,可使组合测度模型克服现有模型只能通过最值强制性对投入和产出进行正向调整以适应选择的效率估计模型无法处理非正值的不足,避免了由此产生的效率估计偏误.实证研究部分用这一组合效率测度模型分析了中国省域层次上大中型工业企业的技术研发效率,为效率的客观比较提供了可行的模型方法.
[Abstract]:On the basis of systematically combing the existing three-stage combined efficiency measurement models of adjustable environmental factors or / and statistical noise, the amplitude adjustment measure (RAM) is combined with stochastic frontier analysis (SFA). An improved three-stage combination efficiency measurement model, RAM-SFA-RAM, is constructed, which does not require mandatory adjustment of environmental factors and statistical noise. Compared with the existing models, the new combined measure model not only can objectively set up the nonlinear production function relationship between environmental factors and relaxation variables, but also can reduce the bias of efficiency estimation in estimating the impact and adjustment of input and output. Moreover, it makes full use of the advantage of RAM model to measure inefficiency, which provides a new analysis way for filtering the difference between environmental factors and statistical noise in terms of input and output, and more importantly, it also provides a new way to analyze the difference between environmental factors and statistical noise in terms of input and output at the same time. The selected RAM has the translation invariance, which can make the combination measure model overcome the deficiency that the existing model can only adjust the input and output positively to adapt to the selection of the efficiency estimation model, which can not deal with the non-positive problem. The resulting efficiency bias is avoided. In the part of empirical research, we use the combined efficiency measurement model to analyze the R & D efficiency of large and medium-sized industrial enterprises in the provincial level in China, which provides a feasible model method for the objective comparison of efficiency.
【作者单位】：中国科学院科技政策与管理科学研究所;中国科学院数学与系统科学研究院;中国科学院大学管理学院;
【基金】：国家自然科学基金资助项目(71471170;71103173) 国家软科学研究计划资助项目(2013GXS4B087) 教育部人文社会科学研究资助项目(13YJC790062)
【分类号】：F425;F273.1

【参考文献】