Overall C as a measure of discrimination in survival analysis:: model specific population value and confidence interval estimation
Overall C as a measure of discrimination in survival analysis:: model specific population value and confidence interval estimation
复制标题
DOI：
10.1002/sim.1802
复制
发表时间：
2004-07-15
影响因子：
2
通讯作者：
D'Agostino, RB
中科院分区：
文献类型：
--
作者：
Pencina, MJ;D'Agostino, RB
生存分析模型判别能力的评价是一个具有重要理论意义和实际应用价值的问题。然而，这个问题比评估线性或逻辑回归的性能更复杂。生物统计学文献中提出了几种不同的测量方法。在本文中，我们研究的属性的整体C指数引入Harrell作为自然延伸的ROC曲线面积的生存分析。我们开发的整体C指数作为一个参数，描述的性能，一个给定的模型应用到正在考虑的populatior和讨论的统计量作为其样本估计。我们发现的整体C和修改后的肯德尔的tau之间的关系，并构建一个置信区间，我们的措施的基础上，其估计的渐近正态性。然后，我们调查通过模拟的长度和覆盖概率的这个区间。最后，我们提出了一个真实的生活中的例子，评估性能的心脏研究模型。版权所有（C）2004约翰威利父子有限公司。
The assessment of the discrimination ability of a survival analysis model is a problem of considerable theoretical interest and important practical applications. This issue is, however, more complex than evaluating the performance of a linear or logistic regression. Several different measures have been proposed in the biostatistical literature. In this paper we investigate the properties of the overall C index introduced by Harrell as a natural extension of the ROC curve area to survival analysis. We develop the overall C index as a parameter describing the performance of a given model applied to the populatior under consideration and discuss the statistic used as its sample estimate. We discover a relationship between the overall C and the modified Kendall's tau and construct a confidence interval for our measure based on the asymptotic normality of its estimate. Then we investigate via simulations the length and coverage probability of this interval. Finally, we present a real life example evaluating the performance of a Framingham Heart Study model. Copyright (C) 2004 John Wiley Sons, Ltd.