High-dimensional variable selection in non-linear mixed effects models using a stochastic EM spike-and-slab - INRAE - Institut national de recherche pour l’agriculture, l’alimentation et l’environnement
Conference Poster Year : 2022

High-dimensional variable selection in non-linear mixed effects models using a stochastic EM spike-and-slab

Abstract

High-dimensional data, with many more covariates than observations, such as genomic data for example, are now commonly analyzed. However, there are few tools for high-dimensional variable selection when the data are observations collected repeatedly on several individuals, and even fewer when the model is nonlinear. Thus, we develop a high-dimensional covariate selection procedure for nonlinear mixed-effects models that are natural models for analyzing this type of data. More precisely, we propose a spike-and-slab variable selection in which we fit using the stochastic approximation version of EM algorithm. Similarly to lasso regression, the set of relevant covariates is selected by exploring a grid of values for the penalization parameter. The proposed approach is much faster than a classical MCMC algorithm and shows very good selection performances on simulated data. Finally, our methodology is applied on growth biological data for selecting relevant genetic markers.
Fichier principal
Vignette du fichier
Poster_ISBA (5).pdf (1.04 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04248003 , version 1 (18-10-2023)

Identifiers

  • HAL Id : hal-04248003 , version 1

Cite

Marion Naveau, Guillaume Kon Kam King, Laure Sansonnet, Maud Delattre. High-dimensional variable selection in non-linear mixed effects models using a stochastic EM spike-and-slab. ISBA World Meeting, Jun 2022, Montréal, Canada. . ⟨hal-04248003⟩
59 View
25 Download

Share

More