Christophe Genolini1, Amandine Lacombe2, René Écochard3, Fabien Subtil3. 1. Inserm UMR U1027, Research Unit on Perinatal Epidemiology and Childhood Disabilities, Adolescent Health, Université Paul Sabatier, Toulouse III, Toulouse, France; CeRSM (EA 2931), UFR STAPS, Université de Paris Ouest-Nanterre-La défense, 92000 Nanterre, France. Electronic address: christophe.genolini@u-paris10.fr. 2. Inserm UMR U1027, Research Unit on Perinatal Epidemiology and Childhood Disabilities, Adolescent Health, Université Paul Sabatier, Toulouse III, Toulouse, France. 3. Hospices Civils de Lyon, Service de Biostatistique, F-69003 Lyon, France; CNRS, UMR5558, Laboratoire de Biomtrie et Biologie Evolutive, Equipe Biostatistique-Sant, F-69100 Villeurbanne, France.
Abstract
BACKGROUND: Longitudinal studies are those in which the same variable is repeatedly measured at different times. More likely than others, these studies suffer from missing values. Because the missing values may impact the statistical analyses, it is important that they be dealt with properly. METHODS: In this paper, we present "CopyMean", a new method to impute (predict) monotone missing values. We compared its efficiency to sixteen imputation methods dedicated to the treatment of missing values in longitudinal data. All these methods were tested on four datasets, real or artificial, presenting markedly different caracteristics. RESULTS: The analysis showed that CopyMean was more efficient in almost all situations.
BACKGROUND: Longitudinal studies are those in which the same variable is repeatedly measured at different times. More likely than others, these studies suffer from missing values. Because the missing values may impact the statistical analyses, it is important that they be dealt with properly. METHODS: In this paper, we present "CopyMean", a new method to impute (predict) monotone missing values. We compared its efficiency to sixteen imputation methods dedicated to the treatment of missing values in longitudinal data. All these methods were tested on four datasets, real or artificial, presenting markedly different caracteristics. RESULTS: The analysis showed that CopyMean was more efficient in almost all situations.
Authors: Lucia Gonzalez-Buendia; Santiago Delgado-Tirado; M Rosa Sanabria; Itziar Fernandez; Rosa M Coco Journal: BMC Ophthalmol Date: 2017-08-18 Impact factor: 2.209
Authors: Sarah C Conner; Sara Lodi; Kathryn L Lunetta; Juan P Casas; Steven A Lubitz; Patrick T Ellinor; Christopher D Anderson; Qiuxi Huang; Justin Coleman; Wendy B White; Emelia J Benjamin; Ludovic Trinquart Journal: J Am Heart Assoc Date: 2019-08-08 Impact factor: 5.501