基于小波理论的我国省域技术创新能力差异化的实证研究

发布时间：2018-05-02 07:56

本文选题：省域技术创新能力 + 评价指标体系　；参考：《江西财经大学》2014年硕士论文

【摘要】：技术创新是衡量区域经济状况以及区域竞争实力的重要指标。探究省域技术创新能力,对明晰我国各省、经济圈乃至全国的技术创新内在驱动力,考察知识溢出对技术进步的促进作用,从而深度分析区域技术与区域创新的耦联效应,并推进区域产业部门的协调均衡发展,分析技术创新各子系统的动态涨落状况,展望技术创新的未来发展趋势,具有不可或缺的意义。本文提出了一种将小波分析和改进灰熵TOPSIS法相结合的省域技术创新能力评价方法——WP法。首先,对小波分析、改进的灰熵TOPSIS法等进行了详细的理论阐释,总结了各类多属性综合评价方法的优缺点；然后,以省域技术创新能力为研究背景,解析了技术创新能力的涵义,并构建省域技术创新能力评价指标体系；进而,对全国31个省市1995—2011年的时间序列数据进行小波分解,根据小波分析的能量守恒定律,选取小波分解后的能量与指标权重的加权和来表示各省的技术创新能力；第四步,用熵值法、灰熵TOPSIS法和WP法对我国31个省市的技术创新能力进行分析；接着,基于上述结果,对省域技术创新能力进行影响因素分析；最后,从省域经济综合竞争力提升、科技资源投入产出增加和技术创新环境优化的角度,分析如何增强三大经济地带(京津唐、长江三角洲和珠江三角洲)的技术创新能力,并选择以经济集聚区为中心的经济辐射路径。本文得出的结论是：通过检验实证结果与区域创新能力报告综合排名的Spearman等级相关度,表明WP法在揭示省域技术创新的经济圈集聚现象和省域技术创新能力排名的准确性方面明显优于其他几种方法。而准确定位产学研机构的职能,有效利用知识、技术的溢出效应,便于实现技术创新资源禀赋的合理化配置,协调化调配产业间的比例,实现技术创新的高效化,为省域技术创新能力的提升、区域技术创新能力的协同发展、全国技术创新网络的构建提供依据,最终为实现经济的持续稳步发展提供科技支持。本文选取1995—2011年各省市技术创新能力指标的时间序列数据,一改大多数论文中从知识创造、获取能力等方面进行指标设定的方法,从科技资源的投入产出角度构建省域技术创新能力的动态综合评价模型,评价各省的技术创新能力。主要的创新之处为：第一,将常用于图像识别、数据压缩的小波变换用于省域技术创新指标数据的分析,去除经济信号中的异方差性和非平稳性,得到“去噪”以后反映最本质经济波动情况的数据,充分体现小波分解时频局部性的优点。根据小波分解能量守恒定理,用各指标能量的范数表征各省市的技术创新能力。第二,小波分析平滑去噪后的指标数据表达出了各指标所反映的本质信息,把时间维加入各指标的欧氏范数求解过程,选取改进的灰熵TOPSIS法进行指标权重的设定,克服了传统指标权重求解方法的缺点,这使得权重的计算结果更加科学、完善,平滑去噪后数据的加权和更加准确地表征了各省的技术创新能力。
[Abstract]:Technological innovation is an important indicator to measure the regional economic situation and regional competitive strength. To explore the technological innovation capability of provincial region, to clarify the internal driving force of the technological innovation of all provinces, economic circles and even the whole country, to investigate the effect of knowledge spillover on technological progress, and to analyze the coupling effect of regional technology and regional innovation in depth. It is of great significance to promote the coordinated and balanced development of regional industrial sectors, analyze the dynamic fluctuation of the subsystems of technological innovation, and look forward to the future development trend of technological innovation.
This paper puts forward a method of evaluating the technological innovation ability of provincial domain, which combines the wavelet analysis with the improved grey entropy TOPSIS method - WP method. First, the wavelet analysis, the improved grey entropy TOPSIS method and so on are explained in detail, and the advantages and disadvantages of all kinds of multi attribute comprehensive evaluation methods are summarized. Then, the technological innovation ability of the provincial region is studied. In the background, the meaning of technological innovation ability is analyzed, and the evaluation index system of provincial technological innovation ability is constructed. Then, the time series data of 31 provinces and cities in China from 1995 to 2011 are decomposed by wavelet analysis. According to the energy conservation law of wavelet analysis, the weight of energy and index weight after wavelet decomposition is selected to express each province. The fourth step is to analyze the technological innovation ability of 31 provinces and cities in China by entropy method, grey entropy TOPSIS method and WP method. Then, based on the above results, the influence factors are analyzed. Finally, the comprehensive competitiveness of the provincial economy is raised, the input and output of scientific and technological resources and the technological innovation environment are increased. The optimization point of view is to analyze how to enhance the technological innovation ability of the three economic zones (Beijing, Tianjin Tangshan, Yangtze River Delta and Pearl River Delta), and choose the economic radiation path centered on the economic agglomeration area.
The conclusion of this paper is that by examining the Spearman grade correlation of the comprehensive ranking of the empirical results and the regional innovation capability report, it shows that the WP method is obviously superior to the other methods in revealing the economic agglomeration of the provincial technological innovation and the accuracy of the provincial technological innovation capacity. The effective use of knowledge and technology spillover effect makes it easy to realize the rationalization of the resource endowment of technological innovation, coordinate and allocate the proportion of the industry, realize the high efficiency of the technological innovation, and provide the basis for the enhancement of the technological innovation ability of the province, the coordinated development of regional technological innovation ability and the construction of the national technological innovation network. Provide scientific and technological support for the sustained and steady development of the economy.
This paper selects the time series data of the technical innovation capability indexes of provinces and cities from 1995 to 2011, and changes the method of setting index from knowledge creation and acquisition ability in most of the papers, and constructs a dynamic comprehensive evaluation model of provincial technological innovation ability from the angle of input and output of scientific and technological resources, and evaluates the technological innovation ability of each province. The main innovations are as follows: firstly, the wavelet transform, which is used in image recognition and data compression, is used in the analysis of provincial technical innovation index data, to remove the heteroscedasticity and non-stationary in the economic signal, and to obtain the data of the most essential economic fluctuation after "denoising", and fully embody the advantages of the time frequency locality of the wavelet decomposition. According to the energy conservation theorem of wavelet decomposition, the technical innovation ability of various provinces and cities is characterized by the norm of each index energy. Second, the essential information reflected by each index is expressed by the index data of smoothing denoising by wavelet analysis, the time dimension is added to the Euclidean norm of each index to solve the process, and the improved grey entropy TOPSIS method is selected to carry out the index right. The weight setting overcame the shortcomings of the traditional index weight solving method, which made the weight calculation more scientific and perfect, and smoothed the weight of the data after the denoising and characterized the technological innovation ability of the provinces more accurately.

【学位授予单位】：江西财经大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F124.3

【参考文献】