摘要
本文应用时变参数状态空间模型,利用1953~2005年中国宏观经济数据,估计了样本区间内我国的全要素生产率(TFP),并与传统的索洛残差方法的计算结果进行了比较。分析表明:时变参数方法得到的TFP增长率计算结果由于不包含方程误差,比索洛残差方法的结果精确;TFP增长率的变化趋势,基本和GDP的增长趋势相同,只是有所滞后,滞后期一般为一年。
A new method named state space model based on time-varying parameter is adopted in this paper. Macro-economic data from 1952 to 2005 of China are used to estimate the total factor productivity growth rate between the sample intervals. Compared with the result of the traditional Solow's Residual Method, the main conclusions are as follows : (1) Due to not in cluding the equation error, the result of total factor productivity growth rate calculated by time-varying parameter method is more precise than that of Solow's Residual Method. (2) The trend of to- tal factor productivity growth rate is basically the same as that of GDP growth, but the former lags the latter.
出处
《数量经济技术经济研究》
CSSCI
北大核心
2008年第2期100-109,121,共11页
Journal of Quantitative & Technological Economics
关键词
全要素生产率
时变参数
状态空间模型
卡尔曼滤波
Total Factor Productivity
Time-varying Parameter
State Space Model
Kalman Filter