A new universal resample-stable bootstrap-based stopping criterion for PLS component construction
| Autor: | Frédéric Bertrand, Myriam Maumy-Bertrand, Nicolas Meyer, Jérémy Magnanensi |
|---|---|
| Přispěvatelé: | Institut de Recherche Mathématique Avancée (IRMA), Université de Strasbourg (UNISTRA)-Centre National de la Recherche Scientifique (CNRS), Laboratoire de biostatistique, CHU Strasbourg, Progression tumorale et microenvironnement. Approches translationnelles et épidémiologie, Université de Strasbourg (UNISTRA)-CHU Strasbourg-Les Hôpitaux Universitaires de Strasbourg (HUS)-Institut Régional du Cancer-Centre Paul Strauss : Centre Régional de Lutte contre le Cancer (CRLCC), Labex IRMIA |
| Rok vydání: | 2016 |
| Předmět: |
0301 basic medicine
Statistics and Probability Poisson distribution 01 natural sciences Theoretical Computer Science 010104 statistics & probability 03 medical and health sciences symbols.namesake 62F40 62F35 PLSR Robustness (computer science) Resampling Latent variable Linear regression Partial least squares regression 0101 mathematics Robustness Statistics - Methodology Mathematics Bootstrap PLSGLR [STAT]Statistics [stat] Statistics - Other Statistics 030104 developmental biology Computational Theory and Mathematics symbols Statistics Probability and Uncertainty [STAT.ME]Statistics [stat]/Methodology [stat.ME] Algorithm |
| Zdroj: | Statistics and Computing. 27:757-774 |
| ISSN: | 1573-1375 0960-3174 |
| Popis: | We develop a new robust stopping criterion in Partial Least Squares Regressions (PLSR) components construction characterised by a high level of stability. This new criterion is defined as a universal one since it is suitable both for PLSR and its extension to Generalized Linear Regressions (PLSGLR). This criterion is based on a non-parametric bootstrap process and has to be computed algorithmically. It allows to test each successive components on a preset significant level alpha. In order to assess its performances and robustness with respect to different noise levels, we perform intensive datasets simulations, with a preset and known number of components to extract, both in the case n>p (n being the number of subjects and p the number of original predictors), and for datasets with n Comment: 31 pages, 20 figures |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink