中国经济文献数据挖掘研究

发布时间：2018-01-22 06:42

本文关键词： 经济文献文本挖掘经济分析方法指标运用数据质量　出处：《广东财经大学》2017年硕士论文　论文类型：学位论文

【摘要】：改革开放以后,中国经济研究取得了长足的发展,积累了大量经济文献成果。然而,从基本理论发展层面来看,无论是微观经济理论,还是宏观经济分析,都毫无疑问地越来越向数量化、模型化的方向发展,经济学在此大趋势中也逐渐演变成为一个庞大的“公理化”的“现代经济学理论体系”,与此同时却忽视了假设的现实合理性、指标基本性质适用性、数据准确性等基础性的问题,由此造成大量论文研究成果与现实经济的脱节,经济研究中许多不足之处,需要我们回过头总结与反思。本文以1979年到2016年的收录于中国知网的中国学术期刊网络出版总库中的8,973,710篇经济学术期刊论文为研究对象,综合的运用文献计量学、文本挖掘技术、统计学方法进行了定量分析和定性分析。从论文数量、年代分布、期刊来源、基金项目概况、关键词等多方面进行了计量分析,在此研究基础上,以经济方法、指标选择、数据质量为切入点,着重分析实证论文普遍忽视的基础性问题。研究得出的主要结论:(1)我国经济领域论文数量呈逐年递增的趋势,但具有阶段性的特征,而这种阶段性特征与经济学科的发展以及现实经济的发展密切相关。另外,经济领域高质量的期刊占比较少,论文整体质量偏低,基金论文量逐年增加,但有96%来自于各级政府机构资助,其他基金项目占比偏少。(2)经济分析方法更多的是建立在对解释对象量化的基础之上,通过“抽象化”方式,运用经济模型对复杂的经济现实进行模拟,且经济模型有越建越复杂趋势,而例如“理论分析”“统计分析”这类基础的方法则沦为复杂分析方法的辅助工具。(3)经济学界对指标运用问题并没有足够的重视,指标运用整体水平偏低,学者们普遍缺乏对指标性质及应用范围的理解,由此造成了指标“误用”、“滥用”以及“少数指标打天下”等的指标运用现状。(4)各政府部门提供的数据“相互打架”或者提供“虚假数据”的现象层出不穷,企业部门“问题数据”也屡禁不止,从而使得部分研究成果失去现实意义。
[Abstract]:Since the reform and opening up, China's economic research has made great progress and accumulated a lot of economic literature. However, from the perspective of basic theory development, whether microeconomic theory or macroeconomic analysis. There is no doubt that more and more quantitative, modelled development of economics in this general trend has gradually evolved into a huge "axiomatic" "modern economic theory system". At the same time, it ignores the basic problems such as the rationality of the hypothesis, the applicability of the basic properties of the indicators, the accuracy of the data, and so on, which leads to the disconnection between the research results of a large number of papers and the real economy. There are many shortcomings in economic research, which need to be summarized and reconsidered. This paper uses 8,973 of the China academic Journals published online from 1979 to 2016 in the General Library of Chinese academic Journals. 710 papers of economic academic journals were analyzed quantitatively and qualitatively by means of bibliometrics, text mining and statistical methods. Fund project overview, key words and other aspects of econometric analysis, on the basis of this research, with economic methods, index selection, data quality as the entry point. The main conclusion of the study is that the number of papers in the economic field of our country is increasing year by year, but it has the characteristic of stage. In addition, the high quality periodicals in the economic field account for less, the overall quality of papers is low, and the number of fund papers is increasing year by year. However, 96% are funded by government agencies at all levels, and other funds account for less. 2) the economic analysis method is more based on the quantification of the explanatory objects through the way of "abstraction". The economic model is used to simulate the complicated economic reality, and the economic model tends to be more and more complex. For example, the basic methods of "theoretical analysis" and "statistical analysis" are reduced to the auxiliary tools of complex analytical methods.) the economists do not pay enough attention to the use of indicators. The overall level of the use of indicators is on the low side, and scholars generally lack the understanding of the nature and scope of application of indicators, resulting in the "misuse" of indicators. "abuse" and "a few indicators to fight the world" and other indicators use the status quo. 4) the data provided by various government departments "fight each other" or provide "false data" phenomenon in endlessly. Enterprise sector "problem data" also repeatedly prohibited, so that some of the research results lost practical significance.
【学位授予单位】：广东财经大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：G353.1;F124

【参考文献】