Skip to Main content Skip to Navigation
Conference papers

Sélection de groupes de variables corrélées par classification ascendante hiérarchique et group-lasso

Quentin Grimonprez 1 Alain Celisse 2, 1 Guillemette Marot 3, 1
1 MODAL - MOdel for Data Analysis and Learning
Inria Lille - Nord Europe, LPP - Laboratoire Paul Painlevé - UMR 8524, METRICS - Evaluation des technologies de santé et des pratiques médicales - ULR 2694, Polytech Lille - École polytechnique universitaire de Lille, Université de Lille, Sciences et Technologies
Abstract : In a context of variable selection, the use of penalized regressions in presence of high correlations might be problematic. Only a subset of the correlated variables is selected. Firstly aggregating related variables can help both for selection and interpretation. However, clustering methods require calibration of additional parameters. We will introduce a new method combining hierarchical clustering and group selection.
Document type :
Conference papers
Complete list of metadata

Cited literature [5 references]  Display  Hide  Download

https://hal.inria.fr/hal-01238248
Contributor : Quentin Grimonprez <>
Submitted on : Friday, December 4, 2015 - 3:30:27 PM
Last modification on : Friday, January 22, 2021 - 1:18:04 PM
Long-term archiving on: : Saturday, April 29, 2017 - 7:31:40 AM

File

sfds.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01238248, version 1

Collections

Citation

Quentin Grimonprez, Alain Celisse, Guillemette Marot. Sélection de groupes de variables corrélées par classification ascendante hiérarchique et group-lasso. 47èmes Journées de Statistique, Jun 2015, Lille, France. ⟨hal-01238248⟩

Share

Metrics

Record views

359

Files downloads

855