技术经济
技術經濟
기술경제
Technology Economics
2014年
3期
103~107
,共null页
企业信用 信用评价 信用风险i分类最优原理
企業信用 信用評價 信用風險i分類最優原理
기업신용 신용평개 신용풍험i분류최우원리
enterprise＇s credit; credit evaluation; credit risk; optimal classification principle
基于分类最优原理进行小企业信用风险评价，即以违约样本和非违约样本的重心距离最大为目标函数，以指标的三角模糊熵及变异系数组成的指标权重区间为约束条件，确定指标的组合权重，评价小企业信用风险状况。应用实例结果表明：利用该方法评价小企业的信用风险，能够更好地区分违约客户与非违约客户，使模型的判别精度有所提高。
基于分類最優原理進行小企業信用風險評價，即以違約樣本和非違約樣本的重心距離最大為目標函數，以指標的三角模糊熵及變異繫數組成的指標權重區間為約束條件，確定指標的組閤權重，評價小企業信用風險狀況。應用實例結果錶明：利用該方法評價小企業的信用風險，能夠更好地區分違約客戶與非違約客戶，使模型的判彆精度有所提高。
기우분류최우원리진행소기업신용풍험평개，즉이위약양본화비위약양본적중심거리최대위목표함수，이지표적삼각모호적급변이계수조성적지표권중구간위약속조건，학정지표적조합권중，평개소기업신용풍험상황。응용실례결과표명：이용해방법평개소기업적신용풍험，능구경호지구분위약객호여비위약객호，사모형적판별정도유소제고。
This paper evaluates the credit risk of small enterprise based on the optimal classification principle. That is,it takes the maximum of distance between default and non-default samples＇ centre gravity as the objective function, and considers the constraint on the range of indicators which are computed through the triangular fuzzy entropy and variable coefficient of the indicator, then calculates the combination weights of indi- cators to evaluate the credit risk of small enterprise. The example result indicates that default and non-default customers could be distinguished better by using this method,and the discrimination accuracy of model could also be improved.
