{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,29]],"date-time":"2026-10-29T02:16:10Z","timestamp":1793240170171,"version":"4.5.0"},"reference-count":47,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2021,6,5]],"date-time":"2021-06-05T00:00:00Z","timestamp":1622851200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,2,1]]},"abstract":"<jats:title>Summary<\/jats:title><jats:p>Functional linear discriminant analysis provides a simple yet efficient method for classification, with the possibility of achieving perfect classification. Several methods have been proposed in the literature that mostly address the dimensionality of the problem. On the other hand, there is growing interest in interpretability of the analysis, which favours a simple and sparse solution. In this paper we propose a new approach that incorporates a type of sparsity that identifies nonzero subdomains in the functional setting, yielding a solution that is easier to interpret without compromising performance. Given the need to embed additional constraints in the solution, we reformulate functional linear discriminant analysis as a regularization problem with an appropriate penalty. Inspired by the success of $\\ell_1$-type regularization at inducing zero coefficients for scalar variables, we develop a new regularization method for functional linear discriminant analysis that incorporates an $L^1$-type penalty, $\\int |f|$, to induce zero regions. We demonstrate that our formulation has a well-defined solution that contains zero regions, achieving functional sparsity in the sense of domain selection. In addition, the misclassification probability of the regularized solution is shown to converge to the Bayes error if the data are Gaussian. Our method does not assume that the underlying function has zero regions in the domain, but it produces a sparse estimator that consistently estimates the true function whether or not the latter is sparse. Using both simulated and real data examples, we demonstrate this property of our method in finite samples through comparisons with existing methods.<\/jats:p>","DOI":"10.1093\/biomet\/asaa107","type":"journal-article","created":{"date-parts":[[2020,12,17]],"date-time":"2020-12-17T17:22:54Z","timestamp":1608225774000},"page":"209-226","source":"Crossref","is-referenced-by-count":24,"title":["Sparse functional linear discriminant analysis"],"prefix":"10.1093","volume":"109","author":[{"given":"Juhyun","family":"Park","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeongyoun","family":"Ahn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongho","family":"Jeon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2021,6,5]]},"reference":[{"key":"2022020204423689800_B1","doi-asserted-by":"crossref","first-page":"1210","DOI":"10.1080\/01621459.2017.1320287","article-title":"On the use of reproducing kernel Hilbert spaces in functional classification","volume":"113","author":"Berrendero,","year":"2018","journal-title":"J. Am. Statist. Assoc."},{"key":"2022020204423689800_B2","doi-asserted-by":"crossref","first-page":"204","DOI":"10.1016\/j.csda.2016.01.019","article-title":"Classification methods for Hilbert data based on surrogate density","volume":"99","author":"Bongiorno,","year":"2016","journal-title":"Comp. Statist. Data Anal."},{"key":"2022020204423689800_B3","doi-asserted-by":"crossref","DOI":"10.1007\/978-0-387-70914-7","volume-title":"Functional Analysis, Sobolev Spaces and Partial Differential Equations","author":"Brezis,","year":"2011"},{"key":"2022020204423689800_B4","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/S0167-7152(99)00036-X","article-title":"Functional linear model","volume":"45","author":"Cardot,","year":"1999","journal-title":"Statist. Prob. Lett."},{"key":"2022020204423689800_B5","first-page":"571","article-title":"Spline estimators for the functional linear model","volume":"13","author":"Cardot,","year":"2003","journal-title":"Statist. Sinica"},{"key":"2022020204423689800_B6","doi-asserted-by":"crossref","first-page":"325","DOI":"10.1007\/s00440-006-0025-2","article-title":"CLT in functional linear regression models","volume":"138","author":"Cardot,","year":"2007","journal-title":"Prob. Theory Rel. Fields"},{"key":"2022020204423689800_B7","doi-asserted-by":"crossref","first-page":"1123","DOI":"10.1093\/ajcn\/40.6.1123","article-title":"A new approach for the estimation of body composition: Infrared interactance","volume":"40","author":"Conway,","year":"1984","journal-title":"Am. J. Clin. Nutr."},{"key":"2022020204423689800_B8","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1214\/07-AOS563","article-title":"Smoothing spline estimators for functional linear regression","volume":"37","author":"Crambes,","year":"2009","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B9","first-page":"545","article-title":"Optimal Bayes classifiers for functional data and density ratios","volume":"104","author":"Dai,","year":"2017","journal-title":"Biometrika"},{"key":"2022020204423689800_B10","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-13395-9","volume-title":"Numerical PDE-Constrained Optimization","author":"De los Reyes,","year":"2015"},{"key":"2022020204423689800_B11","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13065-017-0257-5","article-title":"Non-destructive determination of ethanol levels in fermented alcoholic beverages using Fourier transform mid-infrared spectroscopy","volume":"11","author":"Debebe,","year":"2017","journal-title":"Chem. Central J."},{"key":"2022020204423689800_B12","doi-asserted-by":"crossref","first-page":"1171","DOI":"10.1214\/09-AOS741","article-title":"Defining probability density for a distribution of random functions","volume":"38","author":"Delaigle,","year":"2010","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B13","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/j.1467-9868.2011.01003.x","article-title":"Achieving near perfect classification for functional data","volume":"74","author":"Delaigle,","year":"2012","journal-title":"J. R. Statist. Soc. B"},{"key":"2022020204423689800_B14","doi-asserted-by":"crossref","first-page":"397","DOI":"10.1080\/10618600.1998.10474784","article-title":"Penalized regressions: The bridge versus the lasso","volume":"7","author":"Fu,","year":"2012","journal-title":"J. Comp. Graph. Statist."},{"key":"2022020204423689800_B15","first-page":"194","article-title":"Inequalities for the norms of a function and its derivatives in metric ${L_p}$","volume":"1","author":"Gabushin,","year":"1967","journal-title":"Math. Not. Acad. Sci. USSR"},{"key":"2022020204423689800_B16","doi-asserted-by":"crossref","first-page":"281","DOI":"10.1080\/00401706.2014.902774","article-title":"The Mahalanobis distance for functional data with applications to classification","volume":"57","author":"Galeano,","year":"2015","journal-title":"Technometrics"},{"key":"2022020204423689800_B17","first-page":"210","article-title":"Nonparametric regression analysis of growth curves","volume":"12","author":"Gasser,","year":"1984","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B18","doi-asserted-by":"crossref","first-page":"278","DOI":"10.1016\/j.jmva.2018.09.011","article-title":"Sparse quadratic classification rules via linear dimension reduction","volume":"169","author":"Gaynanova,","year":"2019","journal-title":"J. Mult. Anal."},{"key":"2022020204423689800_B19","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-662-12613-4","volume-title":"Numerical Methods for Nonlinear Variational Problems","author":"Glowinski,","year":"1984"},{"key":"2022020204423689800_B20","doi-asserted-by":"crossref","first-page":"637","DOI":"10.1111\/rssb.12125","article-title":"Truncated linear models for functional data","volume":"78","author":"Hall,","year":"2016","journal-title":"J. R. Statist. Soc. B"},{"key":"2022020204423689800_B21","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1198\/00401700152404273","article-title":"A functional data-analytic approach to signal discrimination","volume":"43","author":"Hall,","year":"2001","journal-title":"Technometrics"},{"key":"2022020204423689800_B22","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1214\/aos\/1176324456","article-title":"Penalized discriminant analysis","volume":"23","author":"Hastie,","year":"1995","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B23","doi-asserted-by":"crossref","DOI":"10.1007\/978-0-387-84858-7","volume-title":"The Elements of Statistical Learning: Data Mining, Inference, and Prediction","author":"Hastie,","year":"2009","edition":"2nd ed"},{"key":"2022020204423689800_B24","doi-asserted-by":"crossref","DOI":"10.1002\/9781118762547","volume-title":"Theoretical Foundations of Functional Data Analysis, with an Introduction to Linear Operators","author":"Hsing,","year":"2015"},{"key":"2022020204423689800_B25","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1111\/1467-9868.00297","article-title":"Functional linear discriminant analysis for irregularly sampled curves","volume":"63","author":"James,","year":"2001","journal-title":"J. R. Statist. Soc. B"},{"key":"2022020204423689800_B26","doi-asserted-by":"crossref","first-page":"2083","DOI":"10.1214\/08-AOS641","article-title":"Functional linear regression that\u2019s interpretable","volume":"37","author":"James,","year":"2009","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B27","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1016\/S0169-7439(97)00038-5","article-title":"Two data sets of near infrared spectra","volume":"37","author":"Kalivas,","year":"1997","journal-title":"Chemomet. Intel. Lab. Syst."},{"key":"2022020204423689800_B28","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1214\/15-AOS1323","article-title":"Functional linear regression with points of impact","volume":"44","author":"Kneip,","year":"2016","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B29","doi-asserted-by":"crossref","first-page":"161","DOI":"10.1093\/biomet\/asy060","article-title":"Classification of functional fragments by regularized linear classifiers with domain selection","volume":"106","author":"Kraus,","year":"2018","journal-title":"Biometrika"},{"key":"2022020204423689800_B30","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1137\/17S015859","article-title":"${L}^1$ regularization for compact support","volume":"11","author":"Li,","year":"2018","journal-title":"SIAM Undergrad. Res. Online"},{"key":"2022020204423689800_B31","doi-asserted-by":"crossref","first-page":"3321","DOI":"10.1214\/10-AOS813","article-title":"Uniform convergence rates for nonparametric regression and principal component analysis in functional longitudinal data","volume":"38","author":"Li,","year":"2010","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B32","doi-asserted-by":"crossref","first-page":"306","DOI":"10.1080\/10618600.2016.1195273","article-title":"Locally sparse estimator for functional linear regression models","volume":"26","author":"Lin,","year":"2017","journal-title":"J. Comp. Graph. Statist."},{"key":"2022020204423689800_B33","doi-asserted-by":"crossref","first-page":"846","DOI":"10.1111\/biom.12457","article-title":"Interpretable functional principal component analysis","volume":"72","author":"Lin,","year":"2016","journal-title":"Biometrics"},{"key":"2022020204423689800_B34","doi-asserted-by":"crossref","first-page":"146","DOI":"10.1016\/j.ejor.2012.08.017","article-title":"Interpretable support vector machines for functional data","volume":"232","author":"Martin-Barragan,","year":"2014","journal-title":"Eur. J. Oper. Res."},{"key":"2022020204423689800_B35","doi-asserted-by":"crossref","first-page":"223","DOI":"10.1111\/j.1467-9469.2005.00429.x","article-title":"Functional modelling and classification of longitudinal data","volume":"32","author":"M\u00fcller,","year":"2005","journal-title":"Scand. J. Statist."},{"key":"2022020204423689800_B36","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1007\/s11222-018-9806-6","article-title":"Interpretable sparse SIR for functional data","volume":"29","author":"Picheny,","year":"2019","journal-title":"Statist. Comp."},{"key":"2022020204423689800_B37","year":"2022","journal-title":"R: A Language and Environment for Statistical Computing"},{"key":"2022020204423689800_B38","article-title":"Variable selection and estimation in multivariate functional linear regression via the lasso","author":"Roche,","year":"2019"},{"key":"2022020204423689800_B39","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1080\/030144699282642","article-title":"An analysis of variance of the pubertal and midgrowth spurts for length and width","volume":"26","author":"Sheehy,","year":"1999","journal-title":"Ann. Hum. Biol."},{"key":"2022020204423689800_B40","doi-asserted-by":"crossref","first-page":"1191","DOI":"10.1016\/j.jmva.2007.08.001","article-title":"An extension of Fisher\u2019s discriminant analysis for stochastic processes","volume":"99","author":"Shin,","year":"2008","journal-title":"J. Mult. Anal."},{"key":"2022020204423689800_B41","doi-asserted-by":"crossref","first-page":"282","DOI":"10.1016\/j.csda.2012.06.017","article-title":"Interpretable dimension reduction for classifying functional data","volume":"57","author":"Tian,","year":"2013","journal-title":"Comp. Statist. Data Anal."},{"key":"2022020204423689800_B42","first-page":"439","article-title":"Estimation of functional sparsity in nonparametric varying coefficient models for longitudinal data analysis","volume":"29","author":"Tu,","year":"2020","journal-title":"Statist. Sinica"},{"key":"2022020204423689800_B43","first-page":"1337","article-title":"Functional sparsity: Global versus local","volume":"25","author":"Wang,","year":"2015","journal-title":"Statist. Sinica"},{"key":"2022020204423689800_B44","doi-asserted-by":"crossref","first-page":"577","DOI":"10.1198\/016214504000001745","article-title":"Functional data analysis for sparse longitudinal data","volume":"100","author":"Yao,","year":"2005","journal-title":"J. Am. Statist. Assoc."},{"key":"2022020204423689800_B45","doi-asserted-by":"crossref","first-page":"2281","DOI":"10.1214\/16-AOS1446","article-title":"From sparse to dense functional data and beyond","volume":"44","author":"Zhang,","year":"2016","journal-title":"Ann. Statist."},{"key":"2022020204423689800_B46","first-page":"25","article-title":"Functional linear model with zero-value coefficient function at sub-regions","volume":"23","author":"Zhou,","year":"2013","journal-title":"Statist. Sinica"},{"key":"2022020204423689800_B47","doi-asserted-by":"crossref","first-page":"600","DOI":"10.1080\/10618600.2012.679241","article-title":"Wavelet-based LASSO in functional linear regression","volume":"21","author":"Zhou,","year":"2012","journal-title":"J. Comp. Graph. Statist."}],"container-title":["Biometrika"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/biomet\/advance-article-pdf\/doi\/10.1093\/biomet\/asaa107\/39568972\/asaa107.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/109\/1\/209\/42362124\/asaa107.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/biomet\/article-pdf\/109\/1\/209\/42362124\/asaa107.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,19]],"date-time":"2024-08-19T11:12:25Z","timestamp":1724065945000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/biomet\/article\/109\/1\/209\/6064132"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,6,5]]},"references-count":47,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2021,6,5]]},"published-print":{"date-parts":[[2022,2,1]]}},"URL":"https:\/\/doi.org\/10.1093\/biomet\/asaa107","relation":{},"ISSN":["0006-3444","1464-3510"],"issn-type":[{"value":"0006-3444","type":"print"},{"value":"1464-3510","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,3,1]]},"published":{"date-parts":[[2021,6,5]]}}}