A sensitivity analysis and error bounds for the adaptive lasso.

Tathagata Basu; Jochen Einbeck; Matthias Troffaes

A sensitivity analysis and error bounds for the adaptive lasso.

Tathagata Basu, Jochen Einbeck, Matthias Troffaes

Civil And Environmental Engineering

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

Abstract

Sparse regression is an efficient statistical modelling technique which is of major relevance for high dimensional problems. There are several ways of achieving sparse regression, the well-known lasso being one of them. However, lasso variable selection may not be consistent in selecting the true sparse model. Zou (2006) proposed an adaptive form of the lasso which overcomes this issue, and showed that data driven weights on the penalty term will result in a consistent variable selection procedure. Weights can be informed by a prior execution of least squares or ridge regression. Using a power parameter on the weights, we carry out a sensitivity analysis for this parameter, and derive novel error bounds for the Adaptive lasso.

Original language	English
Title of host publication	Proceedings of the 35th International Workshop on Statistical Modelling.
Editors	Itziar Irigoien, Dae-Jin Lee, Joaquín Martínez-Minaya, María Xosé Rodríguez-Álvarez
Pages	278-281
Number of pages	4
Publication status	Published - 20 Jul 2020
Event	35th International Workshop on Statistical Modelling 2020 - Bilbao, Spain Duration: 20 Jul 2020 → 24 Jul 2020

Workshop

Workshop	35th International Workshop on Statistical Modelling 2020
Country/Territory	Spain
City	Bilbao
Period	20/07/20 → 24/07/20

Keywords

sparse regression
statistical modelling technique
adaptive lasso
error bounds
sensitivity analysis

Cite this

@inproceedings{f8443d9cd0394f60bf0ed89c52b90c11,

title = "A sensitivity analysis and error bounds for the adaptive lasso.",

abstract = "Sparse regression is an efficient statistical modelling technique which is of major relevance for high dimensional problems. There are several ways of achieving sparse regression, the well-known lasso being one of them. However, lasso variable selection may not be consistent in selecting the true sparse model. Zou (2006) proposed an adaptive form of the lasso which overcomes this issue, and showed that data driven weights on the penalty term will result in a consistent variable selection procedure. Weights can be informed by a prior execution of least squares or ridge regression. Using a power parameter on the weights, we carry out a sensitivity analysis for this parameter, and derive novel error bounds for the Adaptive lasso.",

keywords = "sparse regression, statistical modelling technique, adaptive lasso, error bounds, sensitivity analysis",

author = "Tathagata Basu and Jochen Einbeck and Matthias Troffaes",

year = "2020",

month = jul,

day = "20",

language = "English",

isbn = "9788413192673",

pages = "278--281",

editor = "Irigoien, {Itziar } and Dae-Jin Lee and Joaqu{\'i}n Mart{\'i}nez-Minaya and Rodr{\'i}guez-{\'A}lvarez, {Mar{\'i}a Xos{\'e}}",

booktitle = "Proceedings of the 35th International Workshop on Statistical Modelling.",

note = "35th International Workshop on Statistical Modelling 2020 ; Conference date: 20-07-2020 Through 24-07-2020",

}

A sensitivity analysis and error bounds for the adaptive lasso. / Basu, Tathagata; Einbeck, Jochen; Troffaes, Matthias.
Proceedings of the 35th International Workshop on Statistical Modelling.. ed. / Itziar Irigoien; Dae-Jin Lee; Joaquín Martínez-Minaya; María Xosé Rodríguez-Álvarez. 2020. p. 278-281.

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

TY - GEN

T1 - A sensitivity analysis and error bounds for the adaptive lasso.

AU - Basu, Tathagata

AU - Einbeck, Jochen

AU - Troffaes, Matthias

PY - 2020/7/20

Y1 - 2020/7/20

N2 - Sparse regression is an efficient statistical modelling technique which is of major relevance for high dimensional problems. There are several ways of achieving sparse regression, the well-known lasso being one of them. However, lasso variable selection may not be consistent in selecting the true sparse model. Zou (2006) proposed an adaptive form of the lasso which overcomes this issue, and showed that data driven weights on the penalty term will result in a consistent variable selection procedure. Weights can be informed by a prior execution of least squares or ridge regression. Using a power parameter on the weights, we carry out a sensitivity analysis for this parameter, and derive novel error bounds for the Adaptive lasso.

AB - Sparse regression is an efficient statistical modelling technique which is of major relevance for high dimensional problems. There are several ways of achieving sparse regression, the well-known lasso being one of them. However, lasso variable selection may not be consistent in selecting the true sparse model. Zou (2006) proposed an adaptive form of the lasso which overcomes this issue, and showed that data driven weights on the penalty term will result in a consistent variable selection procedure. Weights can be informed by a prior execution of least squares or ridge regression. Using a power parameter on the weights, we carry out a sensitivity analysis for this parameter, and derive novel error bounds for the Adaptive lasso.

KW - sparse regression

KW - statistical modelling technique

KW - adaptive lasso

KW - error bounds

KW - sensitivity analysis

M3 - Conference contribution book

SN - 9788413192673

SP - 278

EP - 281

BT - Proceedings of the 35th International Workshop on Statistical Modelling.

A2 - Irigoien, Itziar

A2 - Lee, Dae-Jin

A2 - Martínez-Minaya, Joaquín

A2 - Rodríguez-Álvarez, María Xosé

T2 - 35th International Workshop on Statistical Modelling 2020

Y2 - 20 July 2020 through 24 July 2020

ER -

A sensitivity analysis and error bounds for the adaptive lasso.

Abstract

Workshop

Keywords

Fingerprint

Cite this