| 长三角地区研发企业集聚与知识溢出强度——连续空间中的微观分析 |
| |
| 引用本文： | 王庆喜,胡志学.长三角地区研发企业集聚与知识溢出强度——连续空间中的微观分析[J].地理科学,2018,38(11):1828-1836. |
| |
| 作者姓名： | 王庆喜 胡志学 |
| |
| 作者单位： | 浙江工业大学经贸管理学院, 浙江 杭州 310023 |
| |
| 基金项目： | 国家社会科学基金一般项目（17BJL074）、教育部人文社会科学研究一般项目（15YJA790058）、浙江省哲学社会科学规划课题（18NDJC215YB）资助 |
| |
| 摘 要： | 利用2005年中国规模以上制造业企业数据,以长三角地区为范围,对企业地址进行地理编码,建立企业点对点的空间邻近关系,将距离从5 km逐次扩大到100 km,进行企业知识溢出的空间效应分析。研究发现,长三角地区的研发企业比较集中,大约在20 km范围内有明显的集聚状况,之后则有比较明显的分散,并与整体制造业活动空间分布趋向一致。知识溢出的空间强度随距离递增呈负指数幂函数形式加速衰减,于40 km处左右减势趋于平缓。与连续空间上的微观数据分析相比,区域层面上的归并数据分析放大了知识溢出的效应。
|
| 关 键 词： | 研发企业集聚 DO指数 知识溢出 微观数据分析 |
| 收稿时间： | 2017-11-01 |
| 修稿时间： | 2017-12-19 |
A Micro-level Analysis on R&D Firm Agglomeration and Magnitude of Knowledge Spillovers in Continuous Space |
| |
| Qingxi Wang,Zhixue Hu.A Micro-level Analysis on R&D Firm Agglomeration and Magnitude of Knowledge Spillovers in Continuous Space[J].Scientia Geographica Sinica,2018,38(11):1828-1836. |
| |
| Authors: | Qingxi Wang Zhixue Hu |
| |
| Institution: | School of Economics & Management, Zhejiang University of Technology, Hangzhou 310023, Zhejiang, China |
| |
| Abstract: | Former studies on knowledge spillovers often deal with aggregate data at regional level. However, the observed knowledge spillover comes from spatial interaction among micro entities, and needs to be examined microcosmically in nature. This article, employing micro data of Chinese manufacturing firms above designated size in 2005, analyzes spatial effects of firm knowledge spillovers in the Yangtze River Delta. Firstly, this article identifies geographical coordinates of all firms from Baidu API by R language program, and then calculates the point-to-point geosphere distances among firms. Secondly, with a step of 1 km in a range of 0-100 km, this article explores spatial point pattern of firms with DO index. This article discovers that R&D firms are significantly clustered within a scope of about 20 km, then disperse as distance increases, and tend to accord with the spatial distribution of all the manufacturing firms in the end. This article compares the results with those drawn from EG index calculated at various regional levels under Monte Carlo simulation, and finds that R&D firms are significantly agglomerated at village, town or county levels. However, at city level, firm agglomeration is not significantly observed. The significance level becomes lower as regional level gets higher. Thirdly, this paper constructs spatial contiguity relationships based on threshold distance. To observe how spatial effects of knowledge spillovers vary as distance increases, this paper set multiple threshold distances step by step from 5 km to 100 km, with a step of 5 km. With the package of spdep in R language, this article analyzes the spatial effects of knowledge spillovers at individual level in continuous space based on knowledge production function. This paper finds that, within range of 0-100 km, the spatial effects of firm knowledge spillovers decrease more and more rapidly, and can be described with a fitted function of negative exponential power. Beyond 40 km, the effect of knowledge spillovers moves down a stable low level. This shows that the spatial extent of firm knowledge spillover may be within a distance of about 40 km. Compared with micro data analysis in continuous space, aggregate data analyses based on regional analysis exaggerate the effects of knowledge spillovers. With the analysis of micro data, this article can avoid the modifiable area unit problem effectively, which is a big trouble appearing in the studies based on aggregate data. Under the limitation of the capability of computers, this paper only analyzes the firm data of the Yangtze River Delta in 2005. Subsequent researches can expand the spatial extent and prolong time duration to obtain more general results. |
| |
| Keywords: | R&D firm agglomeration DO index knowledge spillovers micro data analysis |
|
| 点击此处可从《地理科学》浏览原始摘要信息 |
| 点击此处可从《地理科学》下载免费的PDF全文 |
|
