Sparse sufficient dimension reduction for directional regression

Abstract Sufficient dimension reduction has emerged as a powerful tool for extracting meaningful information within high dimensional datasets over the past few decades. These methods aim to reduce the complexity of data by focusing on its most informative components and this allows us to avoid ‘curs...

Full description

Saved in:
Bibliographic Details
Main Authors: Gayun Kwon, Gijeong Noh, Kyongwon Kim
Format: Article
Language:English
Published: SpringerOpen 2025-07-01
Series:Journal of Big Data
Subjects:
Online Access:https://doi.org/10.1186/s40537-025-01219-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849234925873528832
author Gayun Kwon
Gijeong Noh
Kyongwon Kim
author_facet Gayun Kwon
Gijeong Noh
Kyongwon Kim
author_sort Gayun Kwon
collection DOAJ
description Abstract Sufficient dimension reduction has emerged as a powerful tool for extracting meaningful information within high dimensional datasets over the past few decades. These methods aim to reduce the complexity of data by focusing on its most informative components and this allows us to avoid ‘curse of dimensionality’. However, many sufficient dimension reduction methods have challenges because their outcome has the form of the linear combinations of the original predictors. This can make the interpretation of the extracted components quite difficult, particularly when working with a large number of variables. To address this issue, we introduce a sparse sufficient dimension reduction method for directional regression. Our approach converts generalized eigendecomposition to regression type optimization problem with LASSO constraint and this promotes interpretability by generating sparse estimates. Moreover, we provide theoretical support for our proposed method, establishing non-asymptotic oracle inequalities and convergence guarantees for the associated optimization algorithm. We demonstrate the efficacy of our approach by comparing it against existing methods such as non-sparse directional regression, sparse sliced inverse regression, and sliced average variance estimation through comprehensive numerical experiments. We further apply our method to two real-world datasets to present its practical value in extracting meaningful insights from complex data structures.
format Article
id doaj-art-acf25a0509c64c69872ed6f743d55aab
institution Kabale University
issn 2196-1115
language English
publishDate 2025-07-01
publisher SpringerOpen
record_format Article
series Journal of Big Data
spelling doaj-art-acf25a0509c64c69872ed6f743d55aab2025-08-20T04:02:57ZengSpringerOpenJournal of Big Data2196-11152025-07-0112112210.1186/s40537-025-01219-1Sparse sufficient dimension reduction for directional regressionGayun Kwon0Gijeong Noh1Kyongwon Kim2Department of Statistics, Ewha Womans UniversityR&D Center, Hyundai SteelDepartment of Applied Statistics, Department of Statistics and Data Science, Yonsei UniversityAbstract Sufficient dimension reduction has emerged as a powerful tool for extracting meaningful information within high dimensional datasets over the past few decades. These methods aim to reduce the complexity of data by focusing on its most informative components and this allows us to avoid ‘curse of dimensionality’. However, many sufficient dimension reduction methods have challenges because their outcome has the form of the linear combinations of the original predictors. This can make the interpretation of the extracted components quite difficult, particularly when working with a large number of variables. To address this issue, we introduce a sparse sufficient dimension reduction method for directional regression. Our approach converts generalized eigendecomposition to regression type optimization problem with LASSO constraint and this promotes interpretability by generating sparse estimates. Moreover, we provide theoretical support for our proposed method, establishing non-asymptotic oracle inequalities and convergence guarantees for the associated optimization algorithm. We demonstrate the efficacy of our approach by comparing it against existing methods such as non-sparse directional regression, sparse sliced inverse regression, and sliced average variance estimation through comprehensive numerical experiments. We further apply our method to two real-world datasets to present its practical value in extracting meaningful insights from complex data structures.https://doi.org/10.1186/s40537-025-01219-1Sufficient dimension reductionDirectional regressionLASSOHigh dimensional dataMultivariate analysis
spellingShingle Gayun Kwon
Gijeong Noh
Kyongwon Kim
Sparse sufficient dimension reduction for directional regression
Journal of Big Data
Sufficient dimension reduction
Directional regression
LASSO
High dimensional data
Multivariate analysis
title Sparse sufficient dimension reduction for directional regression
title_full Sparse sufficient dimension reduction for directional regression
title_fullStr Sparse sufficient dimension reduction for directional regression
title_full_unstemmed Sparse sufficient dimension reduction for directional regression
title_short Sparse sufficient dimension reduction for directional regression
title_sort sparse sufficient dimension reduction for directional regression
topic Sufficient dimension reduction
Directional regression
LASSO
High dimensional data
Multivariate analysis
url https://doi.org/10.1186/s40537-025-01219-1
work_keys_str_mv AT gayunkwon sparsesufficientdimensionreductionfordirectionalregression
AT gijeongnoh sparsesufficientdimensionreductionfordirectionalregression
AT kyongwonkim sparsesufficientdimensionreductionfordirectionalregression