Rushing Christel, Bulusu Anuradha, Hurwitz Herbert I, Nixon Andrew B, Pang Herbert
Department of Biostatistics and Bioinformatics & Duke Cancer Biostatistics, Duke University School of Medicine, Durham, NC, United States.
Department of Medicine, Duke University School of Medicine, Durham, NC, United States.
Comput Biol Med. 2015 Feb;57:123-9. doi: 10.1016/j.compbiomed.2014.11.015. Epub 2014 Dec 9.
A proper internal validation is necessary for the development of a reliable and reproducible prognostic model for external validation. Variable selection is an important step for building prognostic models. However, not many existing approaches couple the ability to specify the number of covariates in the model with a cross-validation algorithm. We describe a user-friendly SAS macro that implements a score selection method and a leave-one-out cross-validation approach. We discuss the method and applications behind this algorithm, as well as details of the SAS macro.
为了开发用于外部验证的可靠且可重复的预后模型,进行适当的内部验证是必要的。变量选择是构建预后模型的重要一步。然而,现有的方法中没有多少能将指定模型中协变量数量的能力与交叉验证算法相结合。我们描述了一个用户友好的SAS宏,它实现了一种评分选择方法和一种留一法交叉验证方法。我们讨论了该算法背后的方法和应用,以及SAS宏的细节。