论文部分内容阅读
论文结合企业全要素生产率(TFP)估计的理论发展,以中国竞争性行业为例,系统比较不同方法的估计结果。结构估计方法试图打开企业生产经营过程黑箱、充分利用企业行为和决策结构方面的信息,因而成为企业TFP估计最有潜力的发展方向。然而Olley and Pakes(1996)、Levinsohn and Petrin(2003)、Ackerberg et al.(2006)等前期的结构估计方法面临着“共线性”问题的困扰,影响其结果的可靠性。结合了传统指数方法的结构估计结果更为可靠。
Based on the theoretical development of total factor productivity (TFP) estimation in China, this dissertation systematically compares the estimation results of different methods with Chinese competitive industries as an example. The structure estimation method attempts to open the black box of the enterprise production and operation process and make full use of the information about the corporate behavior and decision-making structure, thus becoming the most potential development direction for the TFP estimation. However, the previous structural estimation methods such as Olley and Pakes (1996), Levinsohn and Petrin (2003) and Ackerberg et al. (2006) face the problem of “collinearity ” and affect the reliability of their results. The structure estimation combined with the traditional index method is more reliable.