Shrinkage and model selection with correlated variables via weighted fusion

Zhongyin J Daye, X. Jessie Jeng

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

In this paper, we propose the weighted fusion, a new penalized regression and variable selection method for data with correlated variables. The weighted fusion can potentially incorporate information redundancy among correlated variables for estimation and variable selection. Weighted fusion is also useful when the number of predictors p is larger than the number of observations n. It allows the selection of more than n variables in a motivated way. Real data and simulation examples show that weighted fusion can improve variable selection and prediction accuracy.

Original languageEnglish (US)
Pages (from-to)1284-1298
Number of pages15
JournalComputational Statistics and Data Analysis
Volume53
Issue number4
DOIs
StatePublished - Feb 15 2009
Externally publishedYes

Fingerprint

Shrinkage
Model Selection
Fusion
Fusion reactions
Variable Selection
Penalized Regression
Redundancy
Predictors
Prediction
Simulation

ASJC Scopus subject areas

  • Computational Mathematics
  • Computational Theory and Mathematics
  • Statistics and Probability
  • Applied Mathematics

Cite this

Shrinkage and model selection with correlated variables via weighted fusion. / Daye, Zhongyin J; Jeng, X. Jessie.

In: Computational Statistics and Data Analysis, Vol. 53, No. 4, 15.02.2009, p. 1284-1298.

Research output: Contribution to journalArticle

@article{8621bfa05c6b43029e0f07fbf374b3a1,
title = "Shrinkage and model selection with correlated variables via weighted fusion",
abstract = "In this paper, we propose the weighted fusion, a new penalized regression and variable selection method for data with correlated variables. The weighted fusion can potentially incorporate information redundancy among correlated variables for estimation and variable selection. Weighted fusion is also useful when the number of predictors p is larger than the number of observations n. It allows the selection of more than n variables in a motivated way. Real data and simulation examples show that weighted fusion can improve variable selection and prediction accuracy.",
author = "Daye, {Zhongyin J} and Jeng, {X. Jessie}",
year = "2009",
month = "2",
day = "15",
doi = "10.1016/j.csda.2008.11.007",
language = "English (US)",
volume = "53",
pages = "1284--1298",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",
number = "4",

}

TY - JOUR

T1 - Shrinkage and model selection with correlated variables via weighted fusion

AU - Daye, Zhongyin J

AU - Jeng, X. Jessie

PY - 2009/2/15

Y1 - 2009/2/15

N2 - In this paper, we propose the weighted fusion, a new penalized regression and variable selection method for data with correlated variables. The weighted fusion can potentially incorporate information redundancy among correlated variables for estimation and variable selection. Weighted fusion is also useful when the number of predictors p is larger than the number of observations n. It allows the selection of more than n variables in a motivated way. Real data and simulation examples show that weighted fusion can improve variable selection and prediction accuracy.

AB - In this paper, we propose the weighted fusion, a new penalized regression and variable selection method for data with correlated variables. The weighted fusion can potentially incorporate information redundancy among correlated variables for estimation and variable selection. Weighted fusion is also useful when the number of predictors p is larger than the number of observations n. It allows the selection of more than n variables in a motivated way. Real data and simulation examples show that weighted fusion can improve variable selection and prediction accuracy.

UR - http://www.scopus.com/inward/record.url?scp=58549097639&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58549097639&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2008.11.007

DO - 10.1016/j.csda.2008.11.007

M3 - Article

AN - SCOPUS:58549097639

VL - 53

SP - 1284

EP - 1298

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 4

ER -