The optimal linear combination of multiple predictors under the generalized linear models
The optimal linear combination of multiple predictors under the generalized linear models
复制标题
DOI：
10.1016/j.spl.2009.08.002
复制
发表时间：
2009-11-15
影响因子：
0.8
通讯作者：
Lu, Ying
中科院分区：
文献类型：
--
作者：
Jin, Hua;Lu, Ying
临床医生通常可以对一种疾病进行多种替代诊断测试。重要的是同时使用所有好的诊断预测因子，以建立具有更高统计效用的新预测因子。在二元结果的广义线性模型下，证明了链接函数中多个预测因子的线性组合是最优的，即该组合的受试者工作特征（ROC）曲线下的面积是所有可能的线性组合中最大的。结果被应用于骨质疏松性骨折研究（SOF）的数据分析，并与苏和刘的方法进行比较。(C)2009爱思唯尔有限公司版权所有。
Multiple alternative diagnostic tests for one disease are commonly available to clinicians. It is important to use all the good diagnostic predictors simultaneously to establish a new predictor with higher statistical utility. Under the generalized linear model for binary outcomes, the linear combination of multiple predictors in the link function is proved optimal in the sense that the area under the receiver operating characteristic (ROC) curve of this combination is the largest among all possible linear combinations. The result was applied to analysis of the data from the Study of Osteoporotic Fractures (SOF) in comparison with Su and Liu's approach. (C) 2009 Elsevier B.V. All rights reserved.