Interpretable sparse SIR for functional data

DOI10.1007/S11222-018-9806-6zbMATH Open1430.62079arXiv1606.00614OpenAlexW4297851494WikidataQ130166324 ScholiaQ130166324MaRDI QIDQ61772FDOQ61772

Nathalie Villa-Vialaneix, Victor Picheny, Rémi Servien, Victor Picheny, R. Servien, Nathalie Vialaneix

Publication date: 2 March 2018

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: This work focuses on the issue of variable selection in functional regression. Unlike most work in this framework, our approach does not select isolated points in the definition domain of the predictors, nor does it rely on the expansion of the predictors in a given functional basis. It provides an approach to select full intervals made of consecutive points. This feature improves the interpretability of the estimated coefficients and is desirable in the functional framework for which small shifts are frequent when comparing one predictor (curve) to another. Our method is described in a semiparametric framework based on Sliced Inverse Regression (SIR). SIR is an effective method for dimension reduction of high-dimensional data which computes a linear projection of the predictors in a low-dimensional space, without loss on regression information. We extend the approaches of variable selection developed for multidimensional SIR to select intervals rather than separated evaluation points in the definition domain of the functional predictors. Different and equivalent formulations of SIR are combined in a shrinkage approach with a group-LASSO-like penalty. Finally, a fully automated iterative procedure is also proposed to find the critical (interpretable) intervals. The approach is proved efficient on simulated and real data.

Full work available at URL: https://arxiv.org/abs/1606.00614

Recommendations

• Functional sliced inverse regression analysis
• Random sliced inverse regression
• Simultaneous estimations of optimal directions and optimal transformations for functional data
• Variable selection in semiparametric bi-functional models
• Some extensions of multivariate sliced inverse regression

zbMATH Keywords

Lassofunctional regressioninterval selectionridge regressionSIRsliced inverse regression

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Nonparametric regression and quantile regression (62G08) Statistical ranking and selection procedures (62F07) Ridge regression; shrinkage estimators (Lasso) (62J07)

Cites Work

• A graphical tool for selecting the number of slices and the dimension of the model in SIR and SAVE approaches
• The elements of statistical learning. Data mining, inference, and prediction
• Feature selection for functional data
• Variable selection in infinite-dimensional problems
• Functional linear regression with points of impact
• Functional linear regression that's interpretable
• Fractals with point impact in functional linear regression
• On consistency and sparsity for sliced inverse regression in high dimensions
• Bayesian functional linear regression with sparse step functions
• Testing predictor contributions in sufficient dimension reduction.
• Variable selection for functional regression models via the \(L_1\) regularization
• A semiparametric model for Hilbertian random variables
• Most-predictive design points for functional data predictors
• Title not available (Why is that?)
• A non parametric approach for calibration with functional data
• A Note on Sliced Inverse Regression with Regularizations
• Functional Classification in Hilbert Spaces
• Determining the Dimension in Sliced Inverse Regression and Related Methods
• Generalized Cross-Validation as a Method for Choosing a Good Ridge Parameter
• Sliced Inverse Regression for Dimension Reduction
• Atomic Decomposition by Basis Pursuit
• Determining the Dimensionality in Sliced Inverse Regression
• Title not available (Why is that?)
• Functional sliced inverse regression analysis
• Extending Sliced Inverse Regression
• Theory & Methods: Special Invited Paper: Dimension Reduction and Visualization in Discriminant Analysis (with discussion)
• Sparsity and Smoothness Via the Fused Lasso
• Factor screening by sequential bifurcation
• Bootstrap Testing of the Rank of a Matrix via Least-Squared Constrained Estimation
• Title not available (Why is that?)
• Functional Classification with Margin Conditions
• A note on shrinkage sliced inverse regression
• Multilayer Perceptron with Functional Inputs: an Inverse Regression Approach
• Sliced Inverse Regression with Regularizations
• On Sliced Inverse Regression With High-Dimensional Covariates
• Dimension estimation in sufficient dimension reduction: a unifying approach

Cited In (9)

• An informative prior distribution on functions with application to functional regression
• Bayesian functional linear regression with sparse step functions
• Robust sparse functional regression model
• On optimal regression trees to detect critical intervals for multivariate functional data
• Classification tree algorithm for grouped variables
• Selection of time instants and intervals with support vector regression for multivariate functional data
• Functional sliced inverse regression analysis
• SISIR
• Sparse Functional Boxplots for Multivariate Curves

Uses Software

• R
• e1071
• glmnet
• ElemStatLearn
• edrGraphicalTools

This page was built for publication: Interpretable sparse SIR for functional data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q61772)