Comparing ROC curves derived from regression models.

Comparing ROC curves derived from regression models.
复制标题

DOI:
10.1002/sim.5648
复制
发表时间:
2013-04-30
影响因子:
2
通讯作者:
Begg, Colin B.
Begg, Colin B.
中科院分区:
医学3区
文献类型:
--
作者:
Seshan, Venkatraman E.;Goenen, Mithat;Begg, Colin B.

文献摘要

参考文献

被引文献

相似文献

在构建预测模型时,研究人员经常通过比较从包括新标志物的预测模型生成的ROC曲线与从排除新标志物的模型生成的ROC曲线来评估预测标志物的增量值。许多评论家已经注意到经验上的两个ROC区域的测试往往产生一个非显着的结果时,相应的Wald测试从基本的回归模型是显着的。最近的一篇文章使用模拟表明,广泛使用的ROC面积测试产生非常保守的测试大小和极低的功率。在这篇文章中,我们证明了,无论是测试统计量和它的估计方差严重偏差时,嵌套回归模型的预测作为数据输入的测试,我们详细研究这些问题的原因。虽然可以通过消除这些偏倚的重新检验来创建测试参考分布,但Wald或似然比检验仍然是测试新标记物的增量贡献的首选方法。
In constructing predictive models, investigators frequently assess the incremental value of a predictive marker by comparing the ROC curve generated from the predictive model including the new marker with the ROC curve from the model excluding the new marker. Many commentators have noticed empirically that a test of the two ROC areas often produces a non-significant result when a corresponding Wald test from the underlying regression model is significant. A recent article showed using simulations that the widely-used ROC area test produces exceptionally conservative test size and extremely low power. In this article we demonstrate that both the test statistic and its estimated variance are seriously biased when predictions from nested regression models are used as data inputs for the test, and we examine in detail the reasons for these problems. While it is possible to create a test reference distribution by resampling that removes these biases, Wald or likelihood ratio tests remain the preferred approach for testing the incremental contribution of a new marker.
衡量心血管风险的个体预测因素的影响的进步:重新分类措施的作用。
DOI: 10.7326/0003-4819-150-11-200906020-00007
发表时间: 2009-06-02
影响因子: 39.2
作者:
Cook NR;Ridker PM
通讯作者: Ridker PM
DOI: 10.1093/aje/kwh101
发表时间: 2004-05-01
影响因子: 5
作者:
Pepe, MS;Janes, H;Newcomb, P
通讯作者: Newcomb, P
DOI: 10.1212/wnl.0b013e31820af900
发表时间: 2011-02-08
期刊: NEUROLOGY
影响因子: 9.9
作者:
Roe, C. M.;Fagan, A. M.;Morris, J. C.
通讯作者: Morris, J. C.
DOI: 10.1148/radiology.148.3.6878708
发表时间: 1983-01-01
期刊: RADIOLOGY
影响因子: 19.7
作者:
HANLEY, JA;MCNEIL, BJ
通讯作者: MCNEIL, BJ
DOI: 10.1148/radiol.10100886
发表时间: 2011-04-01
期刊: RADIOLOGY
影响因子: 19.7
作者:
Kwon, Sung Woo;Kim, Young Jin;Chang, Hyuk-Jae
通讯作者: Chang, Hyuk-Jae