Titre :

Model-based Clustering with Missing Not At Random Data

Auteur(s) :

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

Titre de la revue :

Statistics and Computing

Éditeur :

Springer Verlag (Germany)

Date de publication :

2024-06-18

ISSN :

0960-3174

Mot(s)-clé(s) en anglais :

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

Discipline(s) HAL :

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

Résumé en anglais : [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 ...
Lire la suite >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.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Projet ANR :

3IA Côte d'Azur
MUSE

Collections :

• Laboratoire Paul Painlevé - UMR 8524

Source :

Harvested from HAL