Variable selection in functional data classification: a maxima-hunting proposal

DOI10.5705/SS.202014.0014MaRDI QIDQ2796872zbMATH OpenOpenAlexFDO

Authors José R. Berrendero, José L. Torrecilla, Antonio Cuevas

Publication date 30 March 2016

Published in Statistica Sinica (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1309.6697

zbMATH Keywords

variable selectionfunctional data analysisdistance correlationsupervised classification

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Abstract: Variable selection is considered in the setting of supervised binary classification with functional data $X(t),tin[0,1]$. By "variable selection" we mean any dimension-reduction method which leads to replace the whole trajectory $X(t),tin[0,1]$, with a low-dimensional vector $(X(t_1),ldots,X(t_k))$ still keeping a similar classification error. Our proposal for variable selection is based on the idea of selecting the local maxima $(t_1,ldots,t_k)$ of the function ${mathcalV}_X^2(t)={mathcalV}^2(X(t),Y)$, where $mathcalV$ denotes the "distance covariance" association measure for random variables due to Sz'ekely, Rizzo and Bakirov (2007). This method provides a simple natural way to deal with the relevance vs. redundancy trade-off which typically appears in variable selection. This paper includes (a) Some theoretical motivation: a result of consistent estimation on the maxima of ${mathcalV}_X^2$ is shown. We also show different theoretical models for the underlying process $X(t)$ under which the relevant information in concentrated in the maxima of ${mathcalV}_X^2$. (b) An extensive empirical study, including about 400 simulated models and real data examples, aimed at comparing our variable selection method with other standard proposals for dimension reduction.

Recommendations

• Variable selection in classification for multivariate functional data
• The mRMR variable selection method: a comparative study for functional data
• Variable selection in functional additive regression models
• Feature selection for functional data
• Variable selection in functional regression models: a review

Cited in

(22)

• Random forest-based approach for physiological functional variable selection for driver's stress level classification
• Robust sparse functional regression model
• Estimating Knee Movement Patterns of Recreational Runners Across Training Sessions Using Multilevel Functional Regression Models
• An RKHS model for variable selection in functional linear regression
• Dynamic functional variable selection for multimodal mHealth data
• Variable selection in classification for multivariate functional data
• Supervised dimensionality reduction via distance correlation maximization
• Functional-bandwidth kernel for support vector machine with functional data: an alternating optimization algorithm
• Domain selection for Gaussian process data: an application to electrocardiogram signals
• 3D point cloud semantic segmentation through functional data analysis
• Improving Estimation in Functional Linear Regression With Points of Impact: Insights Into Google AdWords
• Variation pattern classification of functional data
• Functional diffusion maps
• A novel framework for joint sparse clustering and alignment of functional data
• On dimension reduction models for functional data
• An introduction to recent advances in high/infinite dimensional statistics
• Feature selection for functional data
• Shape classification based on interpoint distance distributions
• On the Use of Reproducing Kernel Hilbert Spaces in Functional Classification
• The common support function with applications
• The mRMR variable selection method: a comparative study for functional data
• Optimum scale selection for 3D point cloud classification through distance correlation

This page was built for publication: Variable selection in functional data classification: a maxima-hunting proposal

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2796872)