• DocumentCode
    3466302
  • Title

    Research on CHAID Decision Tree Model Based on Rating of China´s Small Enterprises

  • Author

    Zhao, Mingqi ; Sun, Zhijun

  • Author_Institution
    Sch. of Math. Sci., Fudan Univ., Shanghai
  • fYear
    2008
  • fDate
    12-14 Oct. 2008
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    The small enterprise loan business of China has become a problem which has to be settled urgently and thus aroused the research on the small enterprise rating methods. There is a variety of exposure types under the New Basel Agreement, among which the exposure types of company are not applicable in the actual situation of China´s small enterprises. Because of small enterprise scale, low loan line and small risk influence on the single transaction, and risk types having the characteristics of retail-exposure, so the small enterprises can be treated as a category of retail-exposure. Meanwhile, the credit of the small enterprises can be scored by the CHAID classification tree model in the data mining method, so as to reflect the probability of contract breach by the score table and rate the small enterprises. The article is to provide an idea for the methods of rating the small enterprises by realizing the method. The result has shown that this method has certain actual significance.
  • Keywords
    contracts; corporate modelling; decision trees; probability; risk management; Basel Agreement; CHAID classification tree model; CHAID decision tree model; China small enterprises; contract breach; probability; retail exposure; small enterprise loan business; small enterprise rating; small risk influence; Banking; Business; Construction industry; Costs; Data mining; Decision trees; Mathematical model; Portfolios; Risk management; Sun;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Wireless Communications, Networking and Mobile Computing, 2008. WiCOM '08. 4th International Conference on
  • Conference_Location
    Dalian
  • Print_ISBN
    978-1-4244-2107-7
  • Electronic_ISBN
    978-1-4244-2108-4
  • Type

    conf

  • DOI
    10.1109/WiCom.2008.2263
  • Filename
    4680452