期刊
AMERICAN JOURNAL OF EPIDEMIOLOGY
卷 179, 期 6, 页码 764-774出版社
OXFORD UNIV PRESS INC
DOI: 10.1093/aje/kwt312
关键词
angina; stable; imputation; missing data; missingness at random; regression trees; simulation; survival
资金
- United Kingdom National Institute for Health Research [RP-PG-0407-10314]
- Wellcome Trust [086091/Z/08/Z, 0938/30/Z/10/Z]
- Medical Research Council [MR/K006584/1, G0902393, G0900724]
- United Kingdom Biobank
- Farr Institute of Health Informatics Research (Health eResearch Centre Network)
- Medical Research Council
- Arthritis Research UK
- British Heart Foundation
- Cancer Research UK
- Economic and Social Research Council
- Engineering and Physical Sciences Research Council
- National Institute of Health Research
- National Institute for Social Care and Health Research (Welsh Assembly Government)
- Chief Scientist Office (Scottish Government Health Directorates)
- Wellcome Trust
- ESRC [ES/H022252/1, ES/G026300/1] Funding Source: UKRI
- MRC [MC_EX_G0800814, G0900724, G0902393, MR/K02180X/1] Funding Source: UKRI
- Economic and Social Research Council [ES/H022252/1, ES/G026300/1] Funding Source: researchfish
- Medical Research Council [MR/K006584/1, MC_EX_G0800814, G0902393, G0900724, MR/K02180X/1] Funding Source: researchfish
Multivariate imputation by chained equations (MICE) is commonly used for imputing missing data in epidemiologic research. The true imputation model may contain nonlinearities which are not included in default imputation models. Random forest imputation is a machine learning technique which can accommodate nonlinearities and interactions and does not require a particular regression model to be specified. We compared parametric MICE with a random forest-based MICE algorithm in 2 simulation studies. The first study used 1,000 random samples of 2,000 persons drawn from the 10,128 stable angina patients in the CALIBER database (Cardiovascular Disease Research using Linked Bespoke Studies and Electronic Records; 2001-2010) with complete data on all covariates. Variables were artificially made missing at random, and the bias and efficiency of parameter estimates obtained using different imputation methods were compared. Both MICE methods produced unbiased estimates of (log) hazard ratios, but random forest was more efficient and produced narrower confidence intervals. The second study used simulated data in which the partially observed variable depended on the fully observed variables in a nonlinear way. Parameter estimates were less biased using random forest MICE, and confidence interval coverage was better. This suggests that random forest imputation may be useful for imputing complex epidemiologic data sets in which some patients have missing data.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据