arrow
返回

Correcting for Optimistic Prediction in Small Data Sets

delete2014-06-24
delete290
delete
OA
AI
G
Gordon C. S. Smith *
S
Shaun R. Seaman
A
Angela Wood
P
Patrick Royston
I
Ian R. White
DOI:10.1093/aje/kwu140delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
The C statistic is a commonly reported measure of screening test performance. Optimistic estimation of the C statistic is a frequent problem because of overfitting of statistical models in small data sets, and methods exist to correct for this issue. However, many studies do not use such methods, and those that do correct for optimism use diverse methods, some of which are known to be biased. We used clinical data sets (United Kingdom Down syndrome screening data from Glasgow (1991-2003), Edinburgh (1999-2003), and Cambridge (1990-2006), as well as Scottish national pregnancy discharge data (2004-2007)) to evaluate different approaches to adjustment for optimism. We found that sample splitting, cross-validation without replication, and leave-1-out cross-validation produced optimism-adjusted estimates of the C statistic that were biased and/or associated with greater absolute error than other available methods. Cross-validation with replication, bootstrapping, and a new method (leave-pairout cross-validation) all generated unbiased optimism-adjusted estimates of the C statistic and had similar absolute errors in the clinical data set. Larger simulation studies confirmed that all 3 methods performed similarly with 10 or more events per variable, or when the C statistic was 0.9 or greater. However, with lower events per variable or lower C statistics, bootstrapping tended to be optimistic but with lower absolute and mean squared errors than both methods of cross-validation.
Keyword:
logistic models
models
statistical
multivariate analysis
receiver operating characteristic curve
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

American Journal of Epidemiology 封面图
American Journal of Epidemiology
IF:
4.8
论文数:
10.0K
被引数:
3.7W

机构

M
MRC Biostatistics Unit
学者数:
775
论文数: 788
被引数: 3.6K
U
University of Cambridge
学者数:
7.7W
论文数: 7.1W
被引数: 13.7W
引用论文

引用论文

Systematic review of statistical methods used in molecular marker studies in cancer
errCANCER
IF5.1
err2008-03-04
err31
errOAAI
errVickers, Andrew J.; Jang, Kwang; Sargent, Daniel; Lija, Hans; Kattan, Michael W.
err分享
err收藏
err分享
err收藏
Clinical risk prediction for pre-eclampsia in nulliparous women: development of model in international prospective cohort
err2011-04-07
err310
errOAAI
errNorth, Robyn A.; McCowan, Lesley M. E.; Dekker, Gustaaf A.; Poston, Lucilla; Chan, Eliza H. Y.; Stewart, Alistair W.; Black, Michael A.; Taylor, Rennae S.; Walker, James J.; Baker, Philip N.; Kenny, Louise C.
err分享
err收藏
Prognosis and prognostic research: validating a prognostic model预后和预后研究: 验证预后模型
err2009-05-28
err1.1K
PREAI
errAltman, Douglas G.; Vergouwe, Yvonne; Royston, Patrick; Moons, Karel G. M.
err分享
err收藏
Triterpenoid saponins from the stems ofClematis parviloba
err2009-04-01
err0
PREAI
errLi-Hua Yan; Li-Zhen Xu; Jia Lin; Shi-Lin Yang; Yu-Lin Feng
err分享
err收藏
学者 查看更多内容