Model selection for logistic regression via association rules analysis

Pannapa Changpetch, Dennis K.J. Lin

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Interaction is very common in reality, but has received little attention in logistic regression literature. This is especially true for higher-order interactions. In conventional logistic regression, interactions are typically ignored. We propose a model selection procedure by implementing an association rules analysis. We do this by (1) exploring the combinations of input variables which have significant impacts to response (via association rules analysis); (2) selecting the potential (low- and high-order) interactions; (3) converting these potential interactions into new dummy variables; and (4) performing variable selections among all the input variables and the newly created dummy variables (interactions) to build up the optimal logistic regression model. Our model selection procedure establishes the optimal combination of main effects and potential interactions. The comparisons are made through thorough simulations. It is shown that the proposed method outperforms the existing methods in all cases. A real-life example is discussed in detail to demonstrate the proposed method.

Original languageEnglish (US)
Pages (from-to)1415-1428
Number of pages14
JournalJournal of Statistical Computation and Simulation
Volume83
Issue number8
DOIs
StatePublished - Aug 1 2013

Fingerprint

Association rules
Association Rules
Logistic Regression
Model Selection
Logistics
Interaction
Selection Procedures
Higher Order
Logistic Regression Model
Main Effect
Variable Selection
Logistic regression
Model selection
Demonstrate
Simulation

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Modeling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

@article{014528f3db904ae9aa7768d41afa81a8,
title = "Model selection for logistic regression via association rules analysis",
abstract = "Interaction is very common in reality, but has received little attention in logistic regression literature. This is especially true for higher-order interactions. In conventional logistic regression, interactions are typically ignored. We propose a model selection procedure by implementing an association rules analysis. We do this by (1) exploring the combinations of input variables which have significant impacts to response (via association rules analysis); (2) selecting the potential (low- and high-order) interactions; (3) converting these potential interactions into new dummy variables; and (4) performing variable selections among all the input variables and the newly created dummy variables (interactions) to build up the optimal logistic regression model. Our model selection procedure establishes the optimal combination of main effects and potential interactions. The comparisons are made through thorough simulations. It is shown that the proposed method outperforms the existing methods in all cases. A real-life example is discussed in detail to demonstrate the proposed method.",
author = "Pannapa Changpetch and Lin, {Dennis K.J.}",
year = "2013",
month = "8",
day = "1",
doi = "10.1080/00949655.2012.662231",
language = "English (US)",
volume = "83",
pages = "1415--1428",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",
number = "8",

}

Model selection for logistic regression via association rules analysis. / Changpetch, Pannapa; Lin, Dennis K.J.

In: Journal of Statistical Computation and Simulation, Vol. 83, No. 8, 01.08.2013, p. 1415-1428.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Model selection for logistic regression via association rules analysis

AU - Changpetch, Pannapa

AU - Lin, Dennis K.J.

PY - 2013/8/1

Y1 - 2013/8/1

N2 - Interaction is very common in reality, but has received little attention in logistic regression literature. This is especially true for higher-order interactions. In conventional logistic regression, interactions are typically ignored. We propose a model selection procedure by implementing an association rules analysis. We do this by (1) exploring the combinations of input variables which have significant impacts to response (via association rules analysis); (2) selecting the potential (low- and high-order) interactions; (3) converting these potential interactions into new dummy variables; and (4) performing variable selections among all the input variables and the newly created dummy variables (interactions) to build up the optimal logistic regression model. Our model selection procedure establishes the optimal combination of main effects and potential interactions. The comparisons are made through thorough simulations. It is shown that the proposed method outperforms the existing methods in all cases. A real-life example is discussed in detail to demonstrate the proposed method.

AB - Interaction is very common in reality, but has received little attention in logistic regression literature. This is especially true for higher-order interactions. In conventional logistic regression, interactions are typically ignored. We propose a model selection procedure by implementing an association rules analysis. We do this by (1) exploring the combinations of input variables which have significant impacts to response (via association rules analysis); (2) selecting the potential (low- and high-order) interactions; (3) converting these potential interactions into new dummy variables; and (4) performing variable selections among all the input variables and the newly created dummy variables (interactions) to build up the optimal logistic regression model. Our model selection procedure establishes the optimal combination of main effects and potential interactions. The comparisons are made through thorough simulations. It is shown that the proposed method outperforms the existing methods in all cases. A real-life example is discussed in detail to demonstrate the proposed method.

UR - http://www.scopus.com/inward/record.url?scp=84882695796&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882695796&partnerID=8YFLogxK

U2 - 10.1080/00949655.2012.662231

DO - 10.1080/00949655.2012.662231

M3 - Article

AN - SCOPUS:84882695796

VL - 83

SP - 1415

EP - 1428

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

IS - 8

ER -