中国科学技术大学学报 ›› 2018, Vol. 48 ›› Issue (6): 467-476.DOI: 10.3969/j.issn.0253-2778.2018.06.005

• 论著 • 上一篇    下一篇

中小企业综合能力评价研究

杜孝平,赵凯琪,袁伟,杨晴虹,张建伟   

  1. 1.北京航空航天大学软件学院,北京 100191;2.清华大学五道口金融学院,北京 1000191;3.北京航空航天大学经济管理学院,北京 100191
  • 收稿日期:2017-09-25 修回日期:2018-04-10 接受日期:2018-04-10 出版日期:2018-06-30 发布日期:2018-04-10
  • 通讯作者: 赵凯琪
  • 作者简介:杜孝平, 男,博士/教授.研究方向:数据挖掘技术及应用, 大数据处理, 智能交通技术等.E-mail:xpdu@buaa.edu.cn
  • 基金资助:
    作者简介: 杜孝平, 男,博士/教授.研究方向:数据挖掘技术及应用, 大数据处理, 智能交通技术等.E-mail:xpdu@buaa.edu.cn

Comprehensive evaluation of small and medium enterprises

DU Xiaoping , ZHAO Kaiqi, YUAN Wei, YANG Qinghong, ZHANG Jianwei   

  1. 1. School of Software, Beihang University, Beijing 100191; 2. Wudaokou School of Finance, Tsinghua University, Beijing 100191;3. School of Economics and Management, Beihang University, Beijing 100191
  • Received:2017-09-25 Revised:2018-04-10 Accepted:2018-04-10 Online:2018-06-30 Published:2018-04-10

摘要: 为解决以往企业评价模型中各方法评价结果不一致、评价角度不全面的问题,构建了CEM(comprehensive evaluation model)综合评价模型.首先,该模型从成长力、竞争力、融资力、团队力、舆论力、外部力、创新力7个方面建立了评价指标体系,并从子系统内部、各子系统之间、各子系统与总体3个方面进行指标筛选;然后,基于主客观综合赋权的思想,采用3种独立评价算法对企业综合能力进行评价,并运用Kendall-W协和系数法对各评价结果进行一致性检验;接着,分别用算术平均算法和因子分析算法进行组合评价, 并用Spearman等级相关系数法对组合评价结果与原始独立评价结果的相关程度进行检验,选出最优的组合评价模型作为最终的综合能力评价模型;最后,基于中国新三板3 430家中小企业3年(2013-2015)的基本数据,对其进行了综合能力评价的实证研究,并利用样本公司滞后一年(2016)的数据,从企业绩效(净利润增长率、主营业务利润率、主营业务增长率)和股票市值两方面验证了本模型的有效性.

关键词: CEM综合评价模型, 层次分析法, 客观赋权法, 一致性检验, 组合评价算法

Abstract: In order to solve the problem of inconsistent evaluation results and incomplete evaluation aspects in previous enterprise evaluation models, this paper constructs a comprehensive evaluation model(CEM), which exploits the advantages of each algorithm and improves the accuracy of the evaluation results. Firstly, the model establishes the evaluation index system from seven aspects including growth force, competitive force, financing force, team force, opinion force, external force and innovation force, and filters the indicators from three aspects. Secondly, it uses three independent, methods to evaluate enterprises, based on the combination of subjectivity and objectivity, and checks the consistency of these three results by Kendall-W co-ordination coefficient. Thirdly, two algorithms, which are arithmetic average and factor analysis, are used respectively to combine the three results above. Fourthly, the correlation between independent evaluation results and combined evaluation results is computed and the more accurate model is selected by spearman rank correlation coefficient. Finally, the model evaluates 3430 small and medium enterprises from the new over-the-counter market of China, based on their three-year data (from 2013 to 2105) and verify the validity of the result from two aspects, which are performance (net profit growth rate, main business margin, main business growth rate) and stock market value.

Key words: CEM, analytic hierarchy process, objective weighting method, consistency check, combination evaluation algorithm

中图分类号: