当前位置: X-MOL 学术Comput. Stat. Data Anal. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A non-parametric test for comparing conditional ROC curves
Computational Statistics & Data Analysis ( IF 1.5 ) Pub Date : 2021-05-01 , DOI: 10.1016/j.csda.2020.107146
Arís Fanjul-Hevia , Wenceslao González-Manteiga , Juan Carlos Pardo-Fernández

Abstract Comparing the accuracy and the behaviour of different diagnostic procedures is one of the main objectives of the Receiver Operating Characteristic (ROC) curve analysis. Along with the diagnostic variables it is usual to observe other covariates, but that extra information has been hardly ever considered for the comparison of this kind of curves. A new non-parametric test is proposed for the comparison of conditional ROC curves. This test is based on a statistic whose theoretical properties are examined, and a bootstrap mechanism is used to calibrate the test. Simulations are run to analyse the practical performance of the test in terms of level approximation and power. An application to real data is also presented to illustrate the procedure.

中文翻译:

用于比较条件 ROC 曲线的非参数检验

摘要 比较不同诊断程序的准确性和行为是接收者操作特征 (ROC) 曲线分析的主要目标之一。除了诊断变量外,通常还会观察其他协变量,但在比较此类曲线时几乎从未考虑过这些额外信息。提出了一种新的非参数检验来比较条件 ROC 曲线。该测试基于检验其理论特性的统计数据,并使用自举机制来校准测试。运行模拟以分析测试在电平近似和功率方面的实际性能。还提供了对实际数据的应用来说明该过程。
更新日期:2021-05-01
down
wechat
bug