Skip to Main content Skip to Navigation
Conference papers

Prédiction en régression clusterwise PLS

Abstract : Clusterwise linear regression aims at partitioning data sets into clusters characterized by their specific coefficients in a linear regression model. High dimensional data and/or the case of multicollinearity can be handled using clusterwise PLS regression based on components which are linear combinations of the initial predictors. The corresponding local PLS models can be used for prediction purpose once the cluster membership of a future observation is determined. The PLS components are related to the clusters and then may differ from one cluster to another. Therefore they cannot be directly used for the cluster membership determination. We propose to use a discriminant analysis on a selected set of principal components from the PCA of the predictors. The method is illustrated on synthetic data.
Document type :
Conference papers
Complete list of metadata

Cited literature [9 references]  Display  Hide  Download
Contributor : Gilbert Saporta Connect in order to contact the contributor
Submitted on : Friday, March 6, 2020 - 11:00:10 AM
Last modification on : Friday, August 5, 2022 - 2:54:00 PM
Long-term archiving on: : Sunday, June 7, 2020 - 1:21:18 PM


Files produced by the author(s)


  • HAL Id : hal-02500605, version 1



Ndèye Niang, Stéphanie Bougeard, Gilbert Saporta. Prédiction en régression clusterwise PLS. 48 èmes Journées de Statistique, May 2016, Montpellier, France. ⟨hal-02500605⟩



Record views


Files downloads