Optimal Thresholds from Non-Normal Mixture

被引：2

作者：

Hong, Chong Sun ^{[1
]}

Joo, Jae Seon ^{[2
]}

机构：

[1] Sungkyunkwan Univ, Dept Stat, 3-53 Myungryun Dong, Seoul 110745, South Korea

[2] Korean Womens Dev Inst, Stat & Panel Ctr, Seoul, South Korea

来源：

KOREAN JOURNAL OF APPLIED STATISTICS | 2010年 / 23卷 / 05期

关键词：

Credit; default; discriminatory; evaluation; optimal threshold; total accuracy; true rate;

D O I：

10.5351/KJAS.2010.23.5.943

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

From a mixture distribution of the score random variable for credit evaluation, there are many methods of estimating optimal thresholds. Most the research news is based on the assumption of normal distributions. In this paper, we extend non-normal distributions such as Weibull, Logistic and Gamma distributions to estimate an optimal threshold by using a hypotheses test method and other methods maximizing the total accuracy and the true rate. The type I and II errors are obtained and compared with their sums. Finally we discuss their efficiency and derive conclusions for non-normal distributions.

引用

页码：943 / 953

页数：11

共 14 条

[1] BAIRAGI R, 1989, SANKHYA SER B, V51, P263
[2] BERRY MJA, 1999, DATA MINING TECHNIQU
[3] Cost curves: An improved method for visualizing classifier performance
Drummond, Chris
Holte, Robert C.
[J]. MACHINE LEARNING, 2006, 65 (01) : 95 - 130
[4] Engelmann B., 2003, SERIES 2 BANKING FIN
[5] THE MEANING AND USE OF THE AREA UNDER A RECEIVER OPERATING CHARACTERISTIC (ROC) CURVE
HANLEY, JA
MCNEIL, BJ
[J]. RADIOLOGY, 1982, 143 (01) : 29 - 36
[6] Keenan S. C., 2001, RISK, V14, P31
[7] The inconsistency of "optimal" cutpoints obtained using two criteria based on the receiver operating characteristic curve
Perkins, NJ
Schisterman, EF
[J]. AMERICAN JOURNAL OF EPIDEMIOLOGY, 2006, 163 (07) : 670 - 675
[8] Provost F., 1997, Proceedings of the Third International Conference on Knowledge Discovery and Data Mining, P43
[9] Tasche D., 2006, VALIDATION INTERNAL
[10] A balanced accuracy function for epistasis modeling in imbalanced datasets using multifactor dimensionality reduction
Velez, Digna R.
White, Bill C.
Motsinger, Alison A.
Bush, William S.
Ritchie, Marylyn D.
Williams, Scott M.
Moore, Jason H.
[J]. GENETIC EPIDEMIOLOGY, 2007, 31 (04) : 306 - 315

← 1 2 →