Bayesian variable selection for binary response models and direct marketing forecasting

Geng CUI, Man Leung WONG, Guichang ZHANG

Research output: Journal PublicationsJournal Article (refereed)

7 Citations (Scopus)

Abstract

Selecting good variables to build forecasting models is a major challenge for direct marketing given the increasing amount and variety of data. This study adopts the Bayesian variable selection (BVS) using informative priors to select variables for binary response models and forecasting for direct marketing. The variable sets by forward selection and BVS are applied to logistic regression and Bayesian networks. The results of validation using a holdout dataset and the entire dataset suggest that BVS improves the performance of the logistic regression model over the forward selection and full variable sets while Bayesian networks achieve better results using BVS. Thus, Bayesian variable selection can help to select variables and build accurate models using innovative forecasting methods. (C) 2010 Elsevier Ltd. All rights reserved.
Original languageEnglish
Pages (from-to)7656-7662
Number of pages7
JournalExpert Systems with Applications
Volume37
Issue number12
DOIs
Publication statusPublished - 1 Dec 2010

Fingerprint

Marketing
Bayesian networks
Logistics

Keywords

  • Bayesian variable selection
  • binary response models
  • direct marketing
  • distribution of priors
  • forecasting models

Cite this

@article{5030af8a1ebc48a39229c1302d5ca290,
title = "Bayesian variable selection for binary response models and direct marketing forecasting",
abstract = "Selecting good variables to build forecasting models is a major challenge for direct marketing given the increasing amount and variety of data. This study adopts the Bayesian variable selection (BVS) using informative priors to select variables for binary response models and forecasting for direct marketing. The variable sets by forward selection and BVS are applied to logistic regression and Bayesian networks. The results of validation using a holdout dataset and the entire dataset suggest that BVS improves the performance of the logistic regression model over the forward selection and full variable sets while Bayesian networks achieve better results using BVS. Thus, Bayesian variable selection can help to select variables and build accurate models using innovative forecasting methods. (C) 2010 Elsevier Ltd. All rights reserved.",
keywords = "Bayesian variable selection, binary response models, direct marketing, distribution of priors, forecasting models",
author = "Geng CUI and WONG, {Man Leung} and Guichang ZHANG",
year = "2010",
month = "12",
day = "1",
doi = "10.1016/j.eswa.2010.04.077",
language = "English",
volume = "37",
pages = "7656--7662",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Ltd",
number = "12",

}

Bayesian variable selection for binary response models and direct marketing forecasting. / CUI, Geng; WONG, Man Leung; ZHANG, Guichang.

In: Expert Systems with Applications, Vol. 37, No. 12, 01.12.2010, p. 7656-7662.

Research output: Journal PublicationsJournal Article (refereed)

TY - JOUR

T1 - Bayesian variable selection for binary response models and direct marketing forecasting

AU - CUI, Geng

AU - WONG, Man Leung

AU - ZHANG, Guichang

PY - 2010/12/1

Y1 - 2010/12/1

N2 - Selecting good variables to build forecasting models is a major challenge for direct marketing given the increasing amount and variety of data. This study adopts the Bayesian variable selection (BVS) using informative priors to select variables for binary response models and forecasting for direct marketing. The variable sets by forward selection and BVS are applied to logistic regression and Bayesian networks. The results of validation using a holdout dataset and the entire dataset suggest that BVS improves the performance of the logistic regression model over the forward selection and full variable sets while Bayesian networks achieve better results using BVS. Thus, Bayesian variable selection can help to select variables and build accurate models using innovative forecasting methods. (C) 2010 Elsevier Ltd. All rights reserved.

AB - Selecting good variables to build forecasting models is a major challenge for direct marketing given the increasing amount and variety of data. This study adopts the Bayesian variable selection (BVS) using informative priors to select variables for binary response models and forecasting for direct marketing. The variable sets by forward selection and BVS are applied to logistic regression and Bayesian networks. The results of validation using a holdout dataset and the entire dataset suggest that BVS improves the performance of the logistic regression model over the forward selection and full variable sets while Bayesian networks achieve better results using BVS. Thus, Bayesian variable selection can help to select variables and build accurate models using innovative forecasting methods. (C) 2010 Elsevier Ltd. All rights reserved.

KW - Bayesian variable selection

KW - binary response models

KW - direct marketing

KW - distribution of priors

KW - forecasting models

UR - http://commons.ln.edu.hk/sw_master/172

U2 - 10.1016/j.eswa.2010.04.077

DO - 10.1016/j.eswa.2010.04.077

M3 - Journal Article (refereed)

VL - 37

SP - 7656

EP - 7662

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 12

ER -