Mossman D
Division of Forensic Psychiatry, Wright State University School of Medicine, Dayton, OH 45401-0927, USA.
Med Decis Making. 1995 Oct-Dec;15(4):358-66. doi: 10.1177/0272989X9501500406.
The methods most commonly used for analyzing receiver operating characteristic (ROC) data incorporate "binormal" assumptions about the latent frequency distributions of test results. Although these assumptions have proved robust to a wide variety of actual frequency distributions, some data sets do not "fit" the binormal model. In such cases, resampling techniques such as the jackknife and the bootstrap provide versatile, distribution-independent, and more appropriate methods for hypothesis testing. This article describes the application of resampling techniques to ROC data for which the binormal assumptions are not appropriate, and suggests that the bootstrap may be especially helpful in determining confidence intervals from small data samples. The widespread availability of ever-faster computers has made resampling methods increasingly accessible and convenient tools for data analysis.
分析接受者操作特征（ROC）数据时最常用的方法包含了关于测试结果潜在频率分布的“双正态”假设。尽管这些假设已被证明对于各种实际频率分布都具有稳健性，但有些数据集并不“符合”双正态模型。在这种情况下，诸如刀切法和自助法等重采样技术为假设检验提供了通用的、与分布无关且更合适的方法。本文描述了重采样技术在双正态假设不适用的ROC数据中的应用，并表明自助法在从小数据样本确定置信区间时可能特别有用。运算速度越来越快的计算机的广泛普及，使得重采样方法成为越来越容易获取且方便的数据分析工具。
