论文部分内容阅读
多维邻近性是近些年国际学术界在区域创新及产业集群方向新的研究视角。本文首先从多维邻近视角出发探讨了地理邻近、认知邻近对高技术产业集群创新的影响机制,并据此提出4个待验假设;进而以我国国家级软件产业园产业集群为典型案例进行实证分析,并创造性地使用人工神经网络为前导的OLS回归分析方法对待验假设进行双重递进检验。实证结果显示:在高技术产业集群的发展和成熟阶段,地理邻近对集群创新绩效产生负的影响,但负影响递减;认知邻近对集群创新绩效产生正影响;集群外部知识的获取有利于集群创新绩效提升;集群直接创新投入也促进创新绩效的提高,但边际报酬递减。
Multidimensional proximity is a new research perspective of international academia in the field of regional innovation and industrial clusters in recent years. In this paper, we first discuss the influence mechanism of geographical proximity and cognitive proximity on the innovation of high-tech industrial clusters from the perspective of multidimensional proximity and propose four hypotheses to be tested accordingly; and then take the example of industrial cluster of national-level software industrial park in our country as an example Analysis, and creatively using artificial neural networks as the leading OLS regression analysis to test the hypothesis for double progressive test. The empirical results show that: in the development and maturity stages of high-tech industrial clusters, geographical proximity has a negative impact on cluster innovation performance, but the negative impact decreases; Cognitive proximity has a positive impact on cluster innovation performance; external knowledge acquisition is beneficial to the cluster Innovation performance improvement; cluster direct innovation investment also promoted the improvement of innovation performance, but diminishing marginal returns.