The Multiple Facets of Partial Least Squares and Related Methods

1. Front Matter

   Pages i-xiv

   PDF

2. Keynotes

   1. Front Matter

      Pages 1-1

      PDF

   2. Partial Least Squares for Heterogeneous Data

      • Peter Bühlmann

      Pages 3-15

   3. On the PLS Algorithm for Multiple Regression (PLS1)

      • Yoshio Takane, Sébastien Loisel

      Pages 17-28

   4. Extending the Finite Iterative Method for Computing the Covariance Matrix Implied by a Recursive Path Model

      • Zouhair El Hadri, Mohamed Hanafi

      Pages 29-43

   5. Which Resampling-Based Error Estimator for Benchmark Studies? A Power Analysis with Application to PLS-LDA

      • Anne-Laure Boulesteix

      Pages 45-57

   6. Path Directions Incoherence in PLS Path Modeling: A Prediction-Oriented Solution

      • Pasquale Dolce, Vincenzo Esposito Vinzi, Carlo Lauro

      Pages 59-70

3. New Developments in Genomics and Brain Imaging

   1. Front Matter

      Pages 71-71

      PDF

   2. Imaging Genetics with Partial Least Squares for Mixed-Data Types (MiMoPLS)

      • Derek Beaton, Michael Kriegsman, ADNI, Joseph Dunlop, Francesca M. Filbey, Hervé Abdi

      Pages 73-91

   3. PLS and Functional Neuroimaging: Bias and Detection Power Across Different Resampling Schemes

      • Nathan Churchill, Babak Afshin-Pour, Stephen Strother

      Pages 93-102

   4. Estimating and Correcting Optimism Bias in Multivariate PLS Regression: Application to the Study of the Association Between Single Nucleotide Polymorphisms and Multivariate Traits in Attention Deficit Hyperactivity Disorder

      • Erica Cunningham, Antonio Ciampi, Ridha Joober, Aurélie Labbe

      Pages 103-113

   5. Discriminant Analysis for Multiway Data

      • Gisela Lechuga, Laurent Le Brusquet, Vincent Perlbarg, Louis Puybasset, Damien Galanaud, Arthur Tenenhaus

      Pages 115-126

4. New and Alternative Methods for Multitable and Path Analysis

   1. Front Matter

      Pages 127-127

      PDF

   2. Structured Variable Selection for Regularized Generalized Canonical Correlation Analysis

      • Tommy Löfstedt, Fouad Hadj-Selem, Vincent Guillemot, Cathy Philippe, Edouard Duchesnay, Vincent Frouin et al.

      Pages 129-139

   3. Supervised Component Generalized Linear Regression with Multiple Explanatory Blocks: THEME-SCGLR

      • Xavier Bry, Catherine Trottier, Fréderic Mortier, Guillaume Cornu, Thomas Verron

      Pages 141-154

   4. Partial Possibilistic Regression Path Modeling

      • Rosaria Romano, Francesco Palumbo

      Pages 155-168

   5. Assessment and Validation in Quantile Composite-Based Path Modeling

      • Cristina Davino, Vincenzo Esposito Vinzi, Pasquale Dolce

      Pages 169-185

5. Advances in Partial Least Square Regression

   1. Front Matter

      Pages 187-187

      PDF

   2. PLS-Frailty Model for Cancer Survival Analysis Based on Gene Expression Profiles

      • Yi Zhou, Yanan Zhu, Siu-wai Leung

      Pages 189-199

   3. Functional Linear Regression Analysis Based on Partial Least Squares and Its Application

      • Huiwen Wang, Lele Huang

      Pages 201-211

This volume presents state of the art theories, new developments, and important applications of Partial Least Square (PLS) methods. The text begins with the invited communications of current leaders in the field who cover the history of PLS, an overview of methodological issues, and recent advances in regression and multi-block approaches. The rest of the volume comprises selected, reviewed contributions from the 8th International Conference on Partial Least Squares and Related Methods held in Paris, France, on 26-28 May, 2014. They are organized in four coherent sections: 1) new developments in genomics and brain imaging, 2) new and alternative methods for multi-table and path analysis, 3) advances in partial least square regression (PLSR), and  4) partial least square path modeling (PLS-PM) breakthroughs and applications. PLS methods are very versatile methods that are now used in areas as diverse as engineering, life science, sociology, psychology, brain imaging, genomics, andbusiness among both academics and practitioners. The selected chapters here highlight this diversity with applied examples as well as the most recent advances.

• Component Analysis
• Multi-block data analysis
• Partial Least Squares
• Regression
• Regularization
• Structural Equation Modeling

• School of Behavioral and Brain Sciences, The University of Texas at Dallas, Richardson, USA

  Hervé Abdi

• ESSEC Business School, Cergy Pontoise CX, France

  Vincenzo Esposito Vinzi

• CNAM, Paris, USA

  Giorgio Russolillo

• CNAM, Paris Cedex 03, France

  Gilbert Saporta

• NEOMA Business School, Rouen, France

  Laura Trinchera

Hervé Abdi—a former Fulbright scholar and resident of the Paris Institute for Advanced Studies—is professor of cognitive neuroscience at the University of Texas at Dallas. His recent research interests focus on the development of analytical and statistical tools for the analysis of very large data sets as generated, for example, by brain imaging, genetic, genomic, and sensory science research. In addition to Partial Least Squares (PLS) models, he is also particularly interested in variable selection and in multi-modal, multi-way, and multi-block methods. He is the author of more than 250 publications (including 12 books) in these domains. 

Vincenzo Esposito Vinzi—a former President of the International Society for Business and Industrial Statistics (ISBIS)—is professor of statistics and Dean of the Faculty at the ESSEC Business School in Paris and Singapore. His research interests focus mainly on methodological developments in multivariate statistics, namelyfactorial methods, structural equation modeling, PLS  regression, and predictive path modeling that are relevant to various applications in business and industry, sensory data, social surveys, and the measurement of customer satisfaction for commodities and public services.

Giorgio Russolillo is associate professor of statistics at the Conservatoire National des Arts et Métiers (CNAM) of Paris where he is a member of the CEDRIC research group. His research interests involve multivariate data analysis models and methods applied to social and life sciences. Much of his current research focuses on PLS algorithms.

Gilbert Saporta—a former president of the International  Association for Statistical Computing—is emeritus professor of applied  statistics at the Conservatoire National des Arts et Métiers (CNAM) of Paris. His research interests are in exploratory multivariate analysis: functional data analysis including functional PLS, sparse methods, and supervised classification. 

Laura Trinchera is assistant professor of computational statistics at the NEOMA business school (Rouen Campus, France).  Her research interests cover multivariate analyses, structural equation models (SEM), PLS methods. She is also interested in variable selection in regression models.

Policies and ethics