Title :

Model-based Clustering with Missing Not At Random Data

Author(s) :

Sportisse, Aude [Auteur]
Université Côte d'Azur [UniCA]
Modèles et algorithmes pour l’intelligence artificielle [MAASAI]
Marbac, Matthieu [Auteur]
Université de Rennes [UR]
Ecole Nationale de la Statistique et de l'Analyse de l'Information [Bruz] [ENSAI]
Centre National de la Recherche Scientifique [CNRS]
Centre de Recherche en Economie et Statistique [Bruz] [CREST]
Laporte, Fabien [Auteur]
Nantes Université [Nantes Univ]
Centre National de la Recherche Scientifique [CNRS]
ITX-lab unité de recherche de l'institut du thorax UMR1087 UMR6291 [ITX-lab]
Celeux, Gilles [Auteur]
Université Paris-Saclay
Statistique mathématique et apprentissage [CELESTE]
Boyer, Claire [Auteur]
Sorbonne Université [SU]
Laboratoire de Probabilités, Statistique et Modélisation [LPSM (UMR_8001)]
Méthodes numériques pour le problème de Monge-Kantorovich et Applications en sciences sociales [MOKAPLAN]
Josse, Julie [Auteur]
Institut Desbrest de santé publique [IDESP]
Médecine de précision par intégration de données et inférence causale [PREMEDICAL]
Biernacki, Christophe [Auteur]
Université Lille Nord (France)
Centre National de la Recherche Scientifique [CNRS]
MOdel for Data Analysis and Learning [MODAL]

Journal title :

Statistics and Computing

Publisher :

Springer Verlag (Germany)

Publication date :

2024-06-18

ISSN :

0960-3174

English keyword(s) :

Model-based Clustering
Informative Missing Values
EM and Stochastic EM Algorithms
Medical Data

HAL domain(s) :

Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

Model-based unsupervised learning, as any learning task, stalls as soon as missingdata occurs. This is even more true when the missing data are informative, or saidmissing not at random (MNAR). In this paper, we propose ...
Show more >Model-based unsupervised learning, as any learning task, stalls as soon as missingdata occurs. This is even more true when the missing data are informative, or saidmissing not at random (MNAR). In this paper, we propose model-based clustering algorithms designed to handle very general types of missing data, including MNAR data. To do so, we introduce a mixture model for different types of data (continuous, count, categorical and mixed) to jointly model the data distribution and the MNAR mechanism, remaining vigilant to the relative degrees of freedom of each. Several MNAR models are discussed, for which the cause of the missingness can depend on both the values of the missing variable themselves and on the class membership. However, we focus on a specific MNAR model, called MNARz, for which the missingness only depends on the class membership. We first underline its ease of estimation, by showing that the statistical inference can be carried out on the data matrix concatenated with the missing mask considering finally a standard MAR mechanism. Consequently, we propose to perform clustering using the Expectation Maximization algorithm, specially developed for this simplified reinterpretation. Finally, we assess the numerical performances of the proposedmethods on synthetic data and on the real medical registry TraumaBase® aswell.Show less >

Language :

Anglais

Popular science :

Non

ANR Project :

3IA Côte d'Azur
MUSE

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL