Regularized Ordinal Regression and the ordinalNet R Package.

IF 5.4 2区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS
Michael J Wurm, Paul J Rathouz, Bret M Hanlon
{"title":"Regularized Ordinal Regression and the ordinalNet R Package.","authors":"Michael J Wurm, Paul J Rathouz, Bret M Hanlon","doi":"10.18637/jss.v099.i06","DOIUrl":null,"url":null,"abstract":"<p><p>Regularization techniques such as the lasso (Tibshirani 1996) and elastic net (Zou and Hastie 2005) can be used to improve regression model coefficient estimation and prediction accuracy, as well as to perform variable selection. Ordinal regression models are widely used in applications where the use of regularization could be beneficial; however, these models are not included in many popular software packages for regularized regression. We propose a coordinate descent algorithm to fit a broad class of ordinal regression models with an elastic net penalty. Furthermore, we demonstrate that each model in this class generalizes to a more flexible form, that can be used to model either ordered or unordered categorical response data. We call this the <i>elementwise link multinomial-ordinal</i> (ELMO) class, and it includes widely used models such as multinomial logistic regression (which also has an ordinal form) and ordinal logistic regression (which also has an unordered multinomial form). We introduce an elastic net penalty class that applies to either model form, and additionally, this penalty can be used to shrink a non-ordinal model toward its ordinal counterpart. Finally, we introduce the R package <b>ordinalNet</b>, which implements the algorithm for this model class.</p>","PeriodicalId":17237,"journal":{"name":"Journal of Statistical Software","volume":null,"pages":null},"PeriodicalIF":5.4000,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8432594/pdf/nihms-1018361.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Statistical Software","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.18637/jss.v099.i06","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0

Abstract

Regularization techniques such as the lasso (Tibshirani 1996) and elastic net (Zou and Hastie 2005) can be used to improve regression model coefficient estimation and prediction accuracy, as well as to perform variable selection. Ordinal regression models are widely used in applications where the use of regularization could be beneficial; however, these models are not included in many popular software packages for regularized regression. We propose a coordinate descent algorithm to fit a broad class of ordinal regression models with an elastic net penalty. Furthermore, we demonstrate that each model in this class generalizes to a more flexible form, that can be used to model either ordered or unordered categorical response data. We call this the elementwise link multinomial-ordinal (ELMO) class, and it includes widely used models such as multinomial logistic regression (which also has an ordinal form) and ordinal logistic regression (which also has an unordered multinomial form). We introduce an elastic net penalty class that applies to either model form, and additionally, this penalty can be used to shrink a non-ordinal model toward its ordinal counterpart. Finally, we introduce the R package ordinalNet, which implements the algorithm for this model class.

Abstract Image

Abstract Image

Abstract Image

正则化正则回归和 ordinalNet R 软件包。
正则化技术,如 lasso(Tibshirani,1996 年)和 elastic net(Zou 和 Hastie,2005 年),可用于提高回归模型的系数估计和预测准确性,以及进行变量选择。正则回归模型在应用中被广泛使用,正则化的使用可能会带来益处;然而,许多流行的正则化回归软件包并不包含这些模型。我们提出了一种坐标下降算法,用于拟合一大类带有弹性网惩罚的序数回归模型。此外,我们还证明了该类模型中的每个模型都可以推广到一种更灵活的形式,既可以用于有序分类数据建模,也可以用于无序分类响应数据建模。我们将其称为元素链接多叉-序数(ELMO)类,它包括广泛使用的模型,如多叉逻辑回归(也有序数形式)和序数逻辑回归(也有无序多叉形式)。我们介绍了一种适用于任一模型形式的弹性净惩罚类,此外,这种惩罚还可用于将非顺序模型缩减为顺序模型。最后,我们介绍了 R 软件包 ordinalNet,它实现了该模型类的算法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Journal of Statistical Software
Journal of Statistical Software 工程技术-计算机:跨学科应用
CiteScore
10.70
自引率
1.70%
发文量
40
审稿时长
6-12 weeks
期刊介绍: The Journal of Statistical Software (JSS) publishes open-source software and corresponding reproducible articles discussing all aspects of the design, implementation, documentation, application, evaluation, comparison, maintainance and distribution of software dedicated to improvement of state-of-the-art in statistical computing in all areas of empirical research. Open-source code and articles are jointly reviewed and published in this journal and should be accessible to a broad community of practitioners, teachers, and researchers in the field of statistics.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信