Robust estimation of models for longitudinal data with dropouts and outliers

被引:1
作者
Zhang, Yuexia [1 ]
Qin, Guoyou [2 ,3 ]
Zhu, Zhongyi [4 ]
Fu, Bo [5 ]
机构
[1] Univ Toronto, Dept Comp & Math Sci, Toronto, ON, Canada
[2] Fudan Univ, Minist Educ, Sch Publ Hlth, Dept Biostat, Shanghai, Peoples R China
[3] Fudan Univ, Key Lab Publ Hlth Safety, Minist Educ, Shanghai, Peoples R China
[4] Fudan Univ, Dept Stat, Shanghai, Peoples R China
[5] Fudan Univ, Sch Data Sci, Shanghai, Peoples R China
基金
英国医学研究理事会; 中国国家自然科学基金;
关键词
Dropout; longitudinal data; missing at random; outlier; robustness; PARTIAL LINEAR-MODELS; MISSING DATA; REGRESSION-MODELS; SEMIPARAMETRIC REGRESSION; RHEUMATOID-ARTHRITIS; REPEATED OUTCOMES; INFERENCE; EFFICIENT; DISABILITY;
D O I
10.1080/02664763.2020.1845623
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Missing data and outliers usually arise in longitudinal studies. Ignoring the effects of missing data and outliers will make the classical generalized estimating equation approach invalid. The longitudinal cohort study of rheumatoid arthritis patients was designed to investigate whether the Health Assessment Questionnaire score was associated with baseline covariates and changed with time. There exist dropouts and outliers in the data. In order to analyze the data, we develop a robust estimating equation approach. To deal with the responses missing at random, we extend a doubly robust method. To achieve robustness against outliers, we utilize an outlier robust method, which corrects the bias induced by outliers through centralizing the covariate matrix in the estimating equation. The doubly robust method for dropouts is easy to combine with the outlier robust method. The proposed method has the property of robustness in the sense that the proposed estimator is not only doubly robust against model misspecification for dropouts when there is no outlier in the data, but also robust against outliers. Consistency and asymptotic normality of the proposed estimator are established under regularity conditions. A comprehensive simulation study and real data analysis demonstrate that the proposed estimator does have the property of robustness.
引用
收藏
页码:902 / 925
页数:24
相关论文
共 39 条
[1]   Doubly robust estimation in missing data and causal inference models [J].
Bang, H .
BIOMETRICS, 2005, 61 (04) :962-972
[2]  
Bansback N, 2006, J RHEUMATOL, V33, P1503
[3]   MULTIPLE IMPUTATION FOR NONRESPONSE IN SURVEYS - RUBIN,DB [J].
CAMPION, WM .
JOURNAL OF MARKETING RESEARCH, 1989, 26 (04) :485-486
[4]  
Fan J., 2018, Local polynomial modelling and its applications: Monographs on statistics and applied probability
[5]  
Fox J., 2018, An R companion to applied regression
[6]   Intrinsic efficiency and multiple robustness in longitudinal studies with drop-out [J].
Han, Peisong .
BIOMETRIKA, 2016, 103 (03) :683-700
[7]   DOUBLY ROBUST AND LOCALLY EFFICIENT ESTIMATION WITH MISSING OUTCOMES [J].
Han, Peisong ;
Wang, Lu ;
Song, Peter X. -K. .
STATISTICA SINICA, 2016, 26 (02) :691-719
[8]   Multiply Robust Estimation in Regression Analysis With Missing Data [J].
Han, Peisong .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2014, 109 (507) :1159-1173
[9]   Estimation with missing data: beyond double robustness [J].
Han, Peisong ;
Wang, Lu .
BIOMETRIKA, 2013, 100 (02) :417-430
[10]   Robust estimation in generalized partial linear models for clustered data [J].
He, XM ;
Fung, WK ;
Zhu, ZY .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2005, 100 (472) :1176-1184