Functional Principal Component Analysis as an Alternative to Mixed‐Effect Models for Describing Sparse Repeated Measures in Presence of Missing Data - Bordeaux Population Health
Article Dans Une Revue Statistics in Medicine Année : 2024

Functional Principal Component Analysis as an Alternative to Mixed‐Effect Models for Describing Sparse Repeated Measures in Presence of Missing Data

Corentin Ségalas
Catherine Helmer
Robin Genuer
Cécile Proust-Lima

Résumé

ABSTRACT Analyzing longitudinal data in health studies is challenging due to sparse and error‐prone measurements, strong within‐individual correlation, missing data and various trajectory shapes. While mixed‐effect models (MM) effectively address these challenges, they remain parametric models and may incur computational costs. In contrast, functional principal component analysis (FPCA) is a non‐parametric approach developed for regular and dense functional data that flexibly describes temporal trajectories at a potentially lower computational cost. This article presents an empirical simulation study evaluating the behavior of FPCA with sparse and error‐prone repeated measures and its robustness under different missing data schemes in comparison with MM. The results show that FPCA is well‐suited in the presence of missing at random data caused by dropout, except in scenarios involving most frequent and systematic dropout. Like MM, FPCA fails under missing not at random mechanism. The FPCA was applied to describe the trajectories of four cognitive functions before clinical dementia and contrast them with those of matched controls in a case‐control study nested in a population‐based aging cohort. The average cognitive declines of future dementia cases showed a sudden divergence from those of their matched controls with a sharp acceleration 5 to 2.5 years prior to diagnosis.
Fichier principal
Vignette du fichier
Statistics in Medicine - 2024 - S%C3%A9galas - Functional Principal Component Analysis as an Alternative to Mixed%E2%80%90Effect Models.pdf (918.64 Ko) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-04693072 , version 1 (10-09-2024)

Identifiants

Citer

Corentin Ségalas, Catherine Helmer, Robin Genuer, Cécile Proust-Lima. Functional Principal Component Analysis as an Alternative to Mixed‐Effect Models for Describing Sparse Repeated Measures in Presence of Missing Data. Statistics in Medicine, 2024, ⟨10.1002/sim.10214⟩. ⟨hal-04693072⟩

Collections

U1219
0 Consultations
0 Téléchargements

Altmetric

Partager

More