Missing data in longitudinal studies

By Michael J. Daniels, Joseph W. Hogan. 2008

443 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/78ea008fa2bd2a2176f6ac06bb92f447f676c787

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Longitudinal Data Analysis. Handbooks of Modern Statistical Methods (2009, methodology): “Specifically, the standard GEE method yields consistent marginal regression parameter estimators provided the responses are MCAR (see, for example, Laird, 1988).”
Effects of Missing Data Methods in Structural Equation Modeling With Nonnormal Longitudinal Data (2009, methodology): “According to Laird (1988) and Rotnitzky and Wypij (1994), if the model is incorrectly specified, the parameter estimates will be inconsistent unless the missing data mechanism is MCAR.”
kml and kml3d: R Packages to Cluster Longitudinal Data (2015, background): “This is particularly true in longitudinal studies where at least one follow-up assessment is often missing (Little 1993; Hedeker and Gibbons 1997; Mallinckrodt, Lane, Schnell, Peng, and Mancuso 2008; Laird 1988).”
MMRM vs. LOCF: A Comprehensive Comparison Based on Simulation Study and 25 NDA Datasets (2009, background): “Laird ( 1988 ) shows that MCAR and MAR are ignorable missingness, and likelihood-based analyses that ignore the missing data mechanism remain valid.”
Handbook of multilevel analysis (2008, background): “In particular, the latest versions of MLwiN (the successor to ML3; [60]) and HLM [66] have worldwide user bases in the social and biomedical sciences numbering in the thousands, and likelihood-based fitting of at least some multilevel models is also now obtainable in more general-purpose statistical”
Much Ado About Nothing (2007, background): “We also do not consider longitudinal or clustered outcomes, for which other complications arise (e.g. Laird 1988; Robins, Rotnitzky, and Zhao 1995; and Jansen et al. 2006).”
Missing data methods in longitudinal studies: a review (2009, methodology): “We refer the reader to the excellent books by Diggle, Heagerty, Liang, Zeger (2002), Fitzmaurice et al (2004), Verbeke and Molenberghs (2000), Molenberghs and Verbeke (2005), Molenberghs and Verbeke (2005), Molenberghs and Kenward (2007), Daniels and Hogan (2008), Fitzmaurice, Davidian, Verbeke, Mol”
Copy Mean: A New Method to Impute Intermittent Missing Values in Longitudinal Studies (2013, methodology): “However, selection models and pattern-mixture models have been proposed when the data are MNAR or when a sensitivity analysis to this assumption is performed [2,4-7].”
Flexible Imputation of Missing Data (2012, background): “There is an extensive literature which often concentrates on the longitudinal case (Verbeke and Molenberghs, 2000; Molenberghs and Verbeke, 2005; Daniels and Hogan, 2008).”
Covariance Estimation: The GLM and Regularization Perspectives (2011, methodology): “For a review of certain advantages of this approach in the context of longitudinal data and some examples of analysis of such data, see Daniels (2005) and Daniels and Hogan (2008).”
Multilevel Models for Ordinal and Nominal Variables (2008, unknown): “Since estimation of model parameters is based on a full-likelihood approach, the missing data are assumed to be “ignorable” conditional on both the model covariates and the observed responses [60].”
A Randomized, Wait-List Controlled Effectiveness Trial Assessing School-Wide Positive Behavior Support in Elementary Schools (2009, background): “Such an analysis can provide unbiased conclusions even in the face of substantial attrition provided the data were missing at random (Laird, 1988; Nich & Carroll, 1997; Schafer & Graham, 2002), that is, that the missing data do not depend on unobserved determinants of the outcome of interest…”