中文摘要：

目前国际学术界对地理邻近与创新论题的研究囿于静态分析视角。本文从产业集群生命周期视角出发分析地理邻近对集群演化不同阶段创新的动态影响机制，并据此提出理论假说；以我国六大汽车产业集群为研究对象，通过曲线拟合，对集群的动态演化过程进行实证分析，结果表明，目前我国六大汽车产业集群均处在快速成长期，均未到达成熟期；采用动态聚类分析法分别对六大汽车产业集群演化阶段进行划分，不同集群的划分界限不一；在此基础上，建立地理邻近与不同演化阶段集群创新绩效之间的计量模型并进行实证分析，得出以下重要结论：（1）地理邻近对集群演化初期和成长期创新的影响显著为正，成长期的影响效应大于初期；（2）同一演化阶段的不同集群地理邻近影响效应存在显著差异；进而提出针对性政策建议。

英文摘要：

The existing international academic researches on the subject of geographical proximity and innovation have been limited in static analysis perspective. In this paper, we analyze the dynamic effect of geographical proximity on cluster innovation from the perspective of industrial cluster＇s life cycle and put forward some theory hypotheses. We take the six domestic automobile industry clusters as our research objects and empirically ana- lyze the dynamic evolution of the cluster by curve fitting. The results show that the six Chinese major automobile industrial clusters are all in the rapid growth phase, and none has reached the maturity period. The dynamic cluster analysis was used to divide the evolution phase of each automobile industrial cluster. Tt is revealed that different clusters have different stage demarcation. Based on this, we set some econometric models between geographic proximity and cluster innovation performance in different evolution stages to make empirical analysis and finally draw two meaningful conclusions as follows. Firstly, the effect of geographic proximity on cluster innovation is positive both in initial and growth phases, and the effect of growth phase is bigger than that of initial phase. Secondly, there are significant differences in the effects of geographic proximity on different clusters within the same evolution phase. At the end of the paper, the corresponding countermeasures and suggestions are put forward.