共 50 条
Estimation of the ROC curve from the Lehmann family
被引:6
|作者:
Jokiel-Rokita, Alicja
[1
]
Topolnicki, Rafal
[1
]
机构:
[1] Wroclaw Univ Sci & Technol, Fac Pure & Appl Math, Wybrzeze Wyspianskiego 27, PL-50370 Wroclaw, Poland
关键词:
Receiver operating characteristic (ROC) curve;
Bi-Weibull ROC curve;
Area under the curve (AUC);
Semiparametric estimation;
SEMIPARAMETRIC ESTIMATION;
D O I:
10.1016/j.csda.2019.106820
中图分类号:
TP39 [计算机的应用];
学科分类号:
081203 ;
0835 ;
摘要:
A semiparametric model of the ROC curve based on the Lehmann family of distributions is an alternative to the popular binormal model. A special case of this model is the Bi-Weibull model. New estimators of the unknown model parameter and consequently the ROC curve from the Lehmann family are presented, and their properties are proved. The accuracy of the proposed estimators is compared with the accuracy of a known estimator based on the partial likelihood method. The conclusion that some of the new estimators perform generally better than their competitor is made. (C) 2019 Elsevier B.V. All rights reserved.
引用
收藏
页数:11
相关论文