Heckman imputation models for binary or continuous MNAR outcomes and MAR predictors

Jacques-Emmanuel Galimard 1, 2, 3, * Sylvie Chevret 1, 2, 3 Emmanuel Curis 4, 5 Matthieu Resche-Rigon 1, 2, 3
* Corresponding author
1 CRESS - U1153 - Equipe 2 : ECSTRA - Epidémiologie Clinique, STatistique, pour la Recherche en Santé
UPD7 - Université Paris Diderot - Paris 7, CRESS (U1153 / UMR_A_1125 / UMR_S_1153) - Centre de Recherche Épidémiologie et Statistique Sorbonne Paris Cité
5 Laboratoire de biomathématiques [Paris]
BioSDTM - EA 7537 - Biostatistique, traitement et modélisation des données biologiques
Abstract : BACKGROUND: Multiple imputation by chained equations (MICE) requires specifying a suitable conditional imputation model for each incomplete variable and then iteratively imputes the missing values. In the presence of missing not at random (MNAR) outcomes, valid statistical inference often requires joint models for missing observations and their indicators of missingness. In this study, we derived an imputation model for missing binary data with MNAR mechanism from Heckman's model using a one-step maximum likelihood estimator. We applied this approach to improve a previously developed approach for MNAR continuous outcomes using Heckman's model and a two-step estimator. These models allow us to use a MICE process and can thus also handle missing at random (MAR) predictors in the same MICE process. METHODS: We simulated 1000 datasets of 500 cases. We generated the following missing data mechanisms on 30% of the outcomes: MAR mechanism, weak MNAR mechanism, and strong MNAR mechanism. We then resimulated the first three cases and added an additional 30% of MAR data on a predictor, resulting in 50% of complete cases. We evaluated and compared the performance of the developed approach to that of a complete case approach and classical Heckman's model estimates. RESULTS: With MNAR outcomes, only methods using Heckman's model were unbiased, and with a MAR predictor, the developed imputation approach outperformed all the other approaches. CONCLUSIONS: In the presence of MAR predictors, we proposed a simple approach to address MNAR binary or continuous outcomes under a Heckman assumption in a MICE procedure. TRIAL REGISTRATION: ClinicalTrials.gov NCT00799760.
Document type :
Journal articles
Complete list of metadatas

Cited literature [41 references]  Display  Hide  Download

https://www.hal.inserm.fr/inserm-02310563
Contributor : Myriam Bodescot <>
Submitted on : Thursday, October 10, 2019 - 11:47:09 AM
Last modification on : Saturday, November 9, 2019 - 7:52:13 PM

File

s12874-018-0547-1.pdf
Publication funded by an institution

Identifiers

Collections

Citation

Jacques-Emmanuel Galimard, Sylvie Chevret, Emmanuel Curis, Matthieu Resche-Rigon. Heckman imputation models for binary or continuous MNAR outcomes and MAR predictors. BMC Medical Research Methodology, BioMed Central, 2018, 18 (1), pp.90. ⟨10.1186/s12874-018-0547-1⟩. ⟨inserm-02310563⟩

Share

Metrics

Record views

81

Files downloads

77