Model selection using information criteria under a new estimation method

Least squares ratio

Eylem Deniz, Oguz Akbilgic, J. Andrew Howe

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

In this study, we evaluate several forms of both Akaike-type and Information Complexity (ICOMP)-type information criteria, in the context of selecting an optimal subset least squares ratio (LSR) regression model. Our simulation studies are designed to mimic many characteristics present in real data - heavy tails, multicollinearity, redundant variables, and completely unnecessary variables. Our findings are that LSR in conjunction with one of the ICOMP criteria is very good at selecting the true model. Finally, we apply these methods to the familiar body fat data set.

Original languageEnglish (US)
Pages (from-to)2043-2050
Number of pages8
JournalJournal of Applied Statistics
Volume38
Issue number9
DOIs
StatePublished - Sep 1 2011

Fingerprint

Information Criterion
Least Square Method
Model Selection
Least Squares
Multicollinearity
Heavy Tails
Regression Model
Simulation Study
Subset
Evaluate
Model selection
Least square method
Least squares
Information criterion
Model
Context
Form
Simulation study
Heavy tails
Regression model

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Model selection using information criteria under a new estimation method : Least squares ratio. / Deniz, Eylem; Akbilgic, Oguz; Andrew Howe, J.

In: Journal of Applied Statistics, Vol. 38, No. 9, 01.09.2011, p. 2043-2050.

Research output: Contribution to journalArticle

@article{696df0dcb26f4c93a55d24f8b5af552a,
title = "Model selection using information criteria under a new estimation method: Least squares ratio",
abstract = "In this study, we evaluate several forms of both Akaike-type and Information Complexity (ICOMP)-type information criteria, in the context of selecting an optimal subset least squares ratio (LSR) regression model. Our simulation studies are designed to mimic many characteristics present in real data - heavy tails, multicollinearity, redundant variables, and completely unnecessary variables. Our findings are that LSR in conjunction with one of the ICOMP criteria is very good at selecting the true model. Finally, we apply these methods to the familiar body fat data set.",
author = "Eylem Deniz and Oguz Akbilgic and {Andrew Howe}, J.",
year = "2011",
month = "9",
day = "1",
doi = "10.1080/02664763.2010.545111",
language = "English (US)",
volume = "38",
pages = "2043--2050",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",
number = "9",

}

TY - JOUR

T1 - Model selection using information criteria under a new estimation method

T2 - Least squares ratio

AU - Deniz, Eylem

AU - Akbilgic, Oguz

AU - Andrew Howe, J.

PY - 2011/9/1

Y1 - 2011/9/1

N2 - In this study, we evaluate several forms of both Akaike-type and Information Complexity (ICOMP)-type information criteria, in the context of selecting an optimal subset least squares ratio (LSR) regression model. Our simulation studies are designed to mimic many characteristics present in real data - heavy tails, multicollinearity, redundant variables, and completely unnecessary variables. Our findings are that LSR in conjunction with one of the ICOMP criteria is very good at selecting the true model. Finally, we apply these methods to the familiar body fat data set.

AB - In this study, we evaluate several forms of both Akaike-type and Information Complexity (ICOMP)-type information criteria, in the context of selecting an optimal subset least squares ratio (LSR) regression model. Our simulation studies are designed to mimic many characteristics present in real data - heavy tails, multicollinearity, redundant variables, and completely unnecessary variables. Our findings are that LSR in conjunction with one of the ICOMP criteria is very good at selecting the true model. Finally, we apply these methods to the familiar body fat data set.

UR - http://www.scopus.com/inward/record.url?scp=79961138958&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79961138958&partnerID=8YFLogxK

U2 - 10.1080/02664763.2010.545111

DO - 10.1080/02664763.2010.545111

M3 - Article

VL - 38

SP - 2043

EP - 2050

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

IS - 9

ER -