High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers

Zhaohu Fan; Matthew Reimherr

doi:10.1109/ICISCE.2016.93

High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers

Zhaohu Fan, Matthew Reimherr

Statistics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In [10] and [16], we proposed tools for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictor. We call these techniques Function-on-Scalar Lasso (FSL) and Adaptive Function-on-Scalar Lasso(AFSL). A scalar group lasso was used to fit the FSL and AFSL estimates. While this approach works well, we improve it by producing custom ADMM methods which are specifically designed for functional data. We propose this new framework as a computational tool for finding FSL estimates. Through our numerical studies, we demonstrate the computational improvement of our methodology.

Original language	English (US)
Title of host publication	Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016
Editors	Shaozi Li, Yun Cheng, Ying Dai
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	397-399
Number of pages	3
ISBN (Electronic)	9781509025350
DOIs	https://doi.org/10.1109/ICISCE.2016.93
Publication status	Published - Oct 31 2016
Event	3rd International Conference on Information Science and Control Engineering, ICISCE 2016 - Beijing, China Duration: Jul 8 2016 → Jul 10 2016

Publication series

Name	Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016

Other

Other	3rd International Conference on Information Science and Control Engineering, ICISCE 2016
Country	China
City	Beijing
Period	7/8/16 → 7/10/16

Fingerprint

All Science Journal Classification (ASJC) codes

Signal Processing
Control and Systems Engineering
Artificial Intelligence
Information Systems
Computer Networks and Communications
Computer Vision and Pattern Recognition

Cite this

Fan, Z., & Reimherr, M. (2016). High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers. In S. Li, Y. Cheng, & Y. Dai (Eds.), Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016 (pp. 397-399). [7726190] (Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICISCE.2016.93

Fan, Zhaohu ; Reimherr, Matthew. / High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers. Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016. editor / Shaozi Li ; Yun Cheng ; Ying Dai. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 397-399 (Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016).

@inproceedings{9b036cb0d89c44e38a9a6bef6be06103,

title = "High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers",

abstract = "In [10] and [16], we proposed tools for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictor. We call these techniques Function-on-Scalar Lasso (FSL) and Adaptive Function-on-Scalar Lasso(AFSL). A scalar group lasso was used to fit the FSL and AFSL estimates. While this approach works well, we improve it by producing custom ADMM methods which are specifically designed for functional data. We propose this new framework as a computational tool for finding FSL estimates. Through our numerical studies, we demonstrate the computational improvement of our methodology.",

author = "Zhaohu Fan and Matthew Reimherr",

year = "2016",

month = "10",

day = "31",

doi = "10.1109/ICISCE.2016.93",

language = "English (US)",

series = "Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "397--399",

editor = "Shaozi Li and Yun Cheng and Ying Dai",

booktitle = "Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016",

address = "United States",

note = "3rd International Conference on Information Science and Control Engineering, ICISCE 2016 ; Conference date: 08-07-2016 Through 10-07-2016",

}

Fan, Z & Reimherr, M 2016, High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers. in S Li, Y Cheng & Y Dai (eds), Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016., 7726190, Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016, Institute of Electrical and Electronics Engineers Inc., pp. 397-399, 3rd International Conference on Information Science and Control Engineering, ICISCE 2016, Beijing, China, 7/8/16. https://doi.org/10.1109/ICISCE.2016.93

High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers. / Fan, Zhaohu; Reimherr, Matthew.

Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016. ed. / Shaozi Li; Yun Cheng; Ying Dai. Institute of Electrical and Electronics Engineers Inc., 2016. p. 397-399 7726190 (Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers

AU - Fan, Zhaohu

AU - Reimherr, Matthew

PY - 2016/10/31

Y1 - 2016/10/31

N2 - In [10] and [16], we proposed tools for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictor. We call these techniques Function-on-Scalar Lasso (FSL) and Adaptive Function-on-Scalar Lasso(AFSL). A scalar group lasso was used to fit the FSL and AFSL estimates. While this approach works well, we improve it by producing custom ADMM methods which are specifically designed for functional data. We propose this new framework as a computational tool for finding FSL estimates. Through our numerical studies, we demonstrate the computational improvement of our methodology.

AB - In [10] and [16], we proposed tools for simultaneous variable selection and parameter estimation in a functional linear model with a functional outcome and a large number of scalar predictor. We call these techniques Function-on-Scalar Lasso (FSL) and Adaptive Function-on-Scalar Lasso(AFSL). A scalar group lasso was used to fit the FSL and AFSL estimates. While this approach works well, we improve it by producing custom ADMM methods which are specifically designed for functional data. We propose this new framework as a computational tool for finding FSL estimates. Through our numerical studies, we demonstrate the computational improvement of our methodology.

UR - http://www.scopus.com/inward/record.url?scp=85001103584&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85001103584&partnerID=8YFLogxK

U2 - 10.1109/ICISCE.2016.93

DO - 10.1109/ICISCE.2016.93

M3 - Conference contribution

AN - SCOPUS:85001103584

T3 - Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016

SP - 397

EP - 399

BT - Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016

A2 - Li, Shaozi

A2 - Cheng, Yun

A2 - Dai, Ying

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd International Conference on Information Science and Control Engineering, ICISCE 2016

Y2 - 8 July 2016 through 10 July 2016

ER -

Fan Z, Reimherr M. High-Dimensional Function-on-Scale Regression via the Alternating Direction Method of Multipliers. In Li S, Cheng Y, Dai Y, editors, Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 397-399. 7726190. (Proceedings - 2016 3rd International Conference on Information Science and Control Engineering, ICISCE 2016). https://doi.org/10.1109/ICISCE.2016.93

Access to Document

10.1109/ICISCE.2016.93