PLS for Big Data: a unified parallel algorithm for regularised group PLS

DOI10.1214/19-SS125zbMATH Open1431.62249arXiv1702.07066OpenAlexW2971816392MaRDI QIDQ2323935FDOQ2323935

Pierre Lafaye de Micheaux, Benoit Liquet, Matthew William Sutton

Publication date: 13 September 2019

Published in: Statistics Surveys (Search for Journal in Brave)

Abstract: Partial Least Squares (PLS) methods have been heavily exploited to analyse the association between two blocs of data. These powerful approaches can be applied to data sets where the number of variables is greater than the number of observations and in presence of high collinearity between variables. Different sparse versions of PLS have been developed to integrate multiple data sets while simultaneously selecting the contributing variables. Sparse modelling is a key factor in obtaining better estimators and identifying associations between multiple data sets. The cornerstone of the sparsity version of PLS methods is the link between the SVD of a matrix (constructed from deflated versions of the original matrices of data) and least squares minimisation in linear regression. We present here an accurate description of the most popular PLS methods, alongside their mathematical proofs. A unified algorithm is proposed to perform all four types of PLS including their regularised versions. Various approaches to decrease the computation time are offered, and we show how the whole procedure can be scalable to big data sets.

Full work available at URL: https://arxiv.org/abs/1702.07066

Recommendations

• DPLS and PPLS: two PLS algorithms for large data sets
• Partial least squares for heterogeneous data
• Big data and partial least‐squares prediction
• Sparse Partial Least Squares Regression for Simultaneous Dimension Reduction and Variable Selection
• Penalized partial least square applied to structured data

zbMATH Keywords

partial least squaressingular value decompositionhigh dimensional dataLasso penalties

Mathematics Subject Classification ID

Statistical aspects of big data and data science (62R07) Measures of association (correlation, canonical correlation, etc.) (62H20) Ridge regression; shrinkage estimators (Lasso) (62J07)

Cites Work

• The Degrees of Freedom of Partial Least Squares Regression
• A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Sparsity and Smoothness Via the Fused Lasso
• Title not available (Why is that?)
• Model Selection and Estimation in Regression with Grouped Variables
• Title not available (Why is that?)
• The solution path of the generalized lasso
• Canonical ridge and econometrics of joint production
• A significance test for the lasso
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• The Collinearity Problem in Linear Regression. The Partial Least Squares (PLS) Approach to Generalized Inverses
• Title not available (Why is that?)
• Sparse principal component analysis via regularized low rank matrix approximation
• A Generalized Least-Square Matrix Decomposition
• Transposable regularized covariance models with an application to missing data imputation
• Sparse Partial Least Squares Regression for Simultaneous Dimension Reduction and Variable Selection
• Canonical Correlation Analysis: An Overview with Application to Learning Methods
• Sparse partial least squares classification for high dimensional data
• A sparse PLS for variable selection when integrating omics data
• Comparison of PLS algorithms when number of objects is much larger than number of variables
• Adaptive multivariate ridge regression
• Online Principal Component Analysis in High Dimension: Which Algorithm to Choose?
• Regularized partial least squares with an application to NMR spectroscopy
• Title not available (Why is that?)
• Phase Retrieval Using Alternating Minimization
• A split-and-merge approach for singular value decomposition of large-scale matrices
• PLS for Big Data: a unified parallel algorithm for regularised group PLS

Cited In (1)

• PLS for Big Data: a unified parallel algorithm for regularised group PLS

Uses Software

• R
• VGAM
• glasso
• glmnet
• mixOmics
• ElemStatLearn
• plsdof
• MNIST
• irlba
• selectiveInference
• PMA
• covTest
• biglasso
• EMNIST
• GitHub

This page was built for publication: PLS for Big Data: a unified parallel algorithm for regularised group PLS

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2323935)