论文部分内容阅读
本文运用地理加权回归(geographical weighted regression,GWR)方法,对1997-2002年期间中国地级市R&D知识溢出的空间非稳定性进行了实证分析。传统的OLS只是对参数进行“平均”或“全局”估计,不能反映参数在不同空间的空间非稳定性;GWR是一个简单、有效的技术,可以反映参数在不同空间的空间非稳定性。研究结果发现:在对R&D知识生产进行参数估计时,GWR模型与OLS模型有显著的差异;R&D知识生产的不同要素存在空间变异。
In this paper, geographical weighted regression (GWR) is used to empirically analyze the spatial instability of R & D knowledge spillover in China’s prefecture-level cities from 1997 to 2002. The traditional OLS only performs “average” or “global” estimation of the parameters and can not reflect the spatial instability of the parameters in different spaces. The GWR is a simple and effective technique that can reflect the parameters in different spatial non-spatial stability. The results show that there is a significant difference between the GWR model and the OLS model when estimating the parameters of R & D knowledge production. There are spatial variances among different elements of R & D knowledge production.