【作者】 马艳艳；

【导师】 刘凤朝；

【作者基本信息】 大连理工大学 ， 经济系统分析与管理， 2011， 博士

【摘要】 知识经济时代,知识是企业最为重要的战略性资源。然而,由于技术创新过程越来越复杂,企业很难具备技术创新所需的全部知识,因此从外界获取新知识就成为提升企业技术创新能力的关键。大学作为新知识、新技术的创造者,是知识溢出的重要源头,在国家技术创新体系建设中发挥着越来越重要的作用,因此如何利用大学知识溢出提升企业自主创新能力就成为一个备受政府和企业关注的话题。第一,在分析大学知识特性的基础上,阐述大学知识特性与大学知识溢出机制之间的关系,为面向企业的大学知识溢出效应理论模型构建确立假设条件和特征参数。大学隐性知识溢出通常发生在一定的空间范围内,是地理距离的衰减函数,主要通过大学衍生企业形成和人力资本流动两种机制实现；大学显性知识溢出能够突破地理距离的限制,跨越行政区甚至跨越国界,主要通过产学社会关系网络实现。面向企业的大学知识溢出的实现取决于大学知识溢出供给与企业对知识溢出需求的匹配关系,此外还受到科技中介服务体系、大学-企业间社会关系以及外部市场与政策环境的制约。第二,在C-D形式的Griliches-Jaffe知识生产函数的基础上,构建面向企业的大学知识溢出效应基本理论模型,综合运用空间经济学和社会网络理论,构建纳入地理邻近性和关系邻近性变量的大学知识溢出效应扩展模型,揭示空间和网络因素作用下大学知识溢出的内在机理,为测度地理邻近性和关系邻近性对大学知识溢出的实证研究奠定理论基础。第三,对基于地理邻近性的大学知识溢出效应扩展模型进行实证检验,采用MoranI指数与MoranI散点图检验大学知识溢出的空间相关性,建立空间权重矩阵,运用极大似然估计法对基于衍生企业形成和人力资本流动的大学知识溢出效应进行空间计量经济估计,研究结果表明大学衍生企业形成和人力资本流动与企业创新产出呈正相关关系,然而作用并不显著。第四,对基于关系邻近性的大学知识溢出效应扩展模型进行实证检验,在专利信息分析系统获取产学合作的关系数据,构建对称邻接矩阵,绘制产学合作网络图,采用社会网络分析技术测度产学合作网络的结构特征,运用计量经济分析技术对基于产学合作网络的大学知识溢出效应进行实证检验,研究结果表明产学合作网络规模、密度、中心势等与大学知识溢出效应存在正相关关系;由于大学专利被企业引用相关数据难以获取,本文仅以清华大学为例对基于专利引用网络的大学知识溢出效应进行考察,研究结果表明清华大学专利被企业引用网络的小世界特征和无标度特征对于促进大学知识溢出具有重要作用。最后,根据理论与实证分析结果,提出相应的政策建议,企业应充分利用与大学的地理邻近性和关系邻近性获取大学知识溢出,从而提升自主创新能力。更多还原

【Abstract】 Knowledge is the most important strategic resources for the industries in the knowledge economy. Technological innovation process is more complex, so it is difficult for the industries to own all the knowledge required for technological innovation. The acquisition of new knowledge from the outside is the key to enhance industrial technological innovation capacity. Universities are seen as the source of new knowledge and technology. Universities also are the important source of knowledge spillovers and are playing an increasingly important role in the national innovation system. So the use of university knowledge spillovers to enhance the industrial independent innovation capability has become a topic which attracts much more concern of the government and industries.First, the study explains the relationship between the characteristics of university knowledge and university knowledge spillover mechanisms based on the analysis the characteristics of university knowledge. This can establish assumptions and characteristic parameters for the university knowledge spillovers theoretical models. University tacit knowledge spillover usually occurs in a certain space. It is a distance decay function. University tacit knowledge spillover mainly occurs through the formation of university spin-offs and the flows of human capital. University explicit knowledge spillover can overcome the limit of geographic distance. It usually occurs across administrative even across national boundaries and mainly through the university-to-industry social networks. University-to-industry knowledge spillover is restricted by the availability of university knowledge spillover, the industrial demand for university knowledge spillover and the perfection degree of the intermediary service system.Second, the study has built the basic theoretical model of university-to-indstry knowledge spillover effects based on the Griliches-Jaffe knowledge production function in C-D form. The study also has built the extended model of university-to-indstry knowledge spillover effect based on the integrated use of the spatial economics and social network theory to reveal the university-to-indstry knowledge spillover mechanism under the role of space and network factors. The models aim to lay the theoretical foundation so as to measure the effects of geographic proximity and relationship proximity on the university-to-indstry knowledge spillover. Third, the study conducts the empirical tests for the extended model of university-to-industry knowledge spillovers which contains the spatial factors. The spatial correlation of university knowledge spillover is inspected by MoranⅠIndex and MoranⅠScatter Plot. The spatial weight matrix is established. The maximum likelihood estimation method is used to estimate university knowledge spillovers based on the formation of spin-offs and the flow of human capital. The results show that both university spin-offs and human capital flows have the positive correlation with industrial innovation outputs. However the effects are not significant.Fourth, the study conducts the empirical tests for the extended model of university-to-industry knowledge spillovers which contains the network factors. Chinese university-industry cooperation networks were constructed according to the data which were obtained from the Patent Analysis System. University-industry cooperation relational data are used to build symmetric adjacency matrix. Then the study draws the university-industry cooperation network diagram. The social network analysis technique is used to measure the structural character of the university-industry cooperation network. The econometric analysis is used to conduct the empirical test for university-to-industry knowledge spillovers based on industry-university cooperative network. The results show that there are positive correlations between the network size, network density, network centralization and university knowledge spillover effects respectively. The data relating to university patents cited by industrial patents are difficult to obtain. This paper only studies university knowledge spillovers based on Tsinghua University patent citation network. The results show that the small-world and scale-free characteristics plays an important role in university-to-industry knowledge spilloversFinally, the study has obtained the corresponding policy recommendations based on the theoretical and empirical analysis. The industries should make full use of the geographical proximity and relationship proximity to abtain the university knowledge spillovers to enhance the industrial independent innovation capability.更多还原