Research On Mechanism Of Spatial Pattern Evolution Of Innovation Networks In Three Major Urban Agglomerations In China:Multi-dimensional Proximit Perspective

The research on innovation network of urban agglomeration has important practical significance for the realization of regional economic transformation and development,the construction of innovative cities and the promotion of new urbanization.This paper regards Yangtze River Delta,Pearl River Delta,Beijing-Tianjin-Hebei region as the space category,and in view of the data about innovative connection,analyzes the spatial pattern and evolution characteristics of urban agglomeration's innovation network from the overall,local and self network levels.Secondly,the QAP multivariate regression method is used to test the relationship between the adjacent distance attribute matrix and the structure matrix of the innovation network..Finally,regarding the innovative linkages between 48 cities as the individual,and integrating into attribute characteristics,proximity power and network effect factors to a two-sided interaction negative binomial regression model based on panel count data,in order to explore the dynamic mechanism of innovation network's spatial structure and characteristics evolution.The result shows that:(1)Each innovation network shows the trend to the dense "small world" network and has a long tailed distribution characteristics.The hierarchical structure of the central city tends to be complex,and the whole network develops from a single structure dominated by a handful of core cities into a polycentric structure.The central city's self network is transformed into a complex and orderly scale-free network.(2)The institution proximity after adjustment plays a significant positive role in the evolution of innovation networks in a single urban agglomeration.In the innovation network of the Yangtze River Delta,the role of geography proximity is still not to be ignored,and there is a growing trend.At the same time,the positive promotion of cognition proximity is becoming more and more obvious,while the performance of technology proximity is not stable.In the Pearl River Delta's innovation network,geography,technology and cognition proximity factors show significant negative effects only at one time.In the innovation network of Beijing-Tianj in-Hebei region,in addition to institution proximity factors,other factors don't have outstanding performance.(3)Space distance has negative influence on technological innovation cooperation,while border adjacency effect plays a positive role.Cognitivon proximity,technology proximity and innovation linkages show positive "U" and inverted "U"relationships.Network effects and institution proximity also show a positive effect.Compared with the individual characteristics,the influence of network structure characteristics is stronger.The interaction of network feature similarity and spatial distance and cognition proximity has a significant positive effect on inter city collaborative innovation.The interaction between the network structure similarity and the neighboring power does not show a significant positive impact.Based on this,this paper summarizes the significance and shortcomings of the research,and puts forward some policy measures to strengthen the overall,local and node level optimization of the network,and enhance the adjacent level of the different dimensions of the city according to the geographical situation.