Numerical comparisons of two effective methods for mixed complementarity problems

Jein-Shan Chen, Shaohua Pan, Ching Yu Yang

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Recently there have two different effective methods proposed by Kanzow et al. in (Kanzow, 2001 [8]) and (Kanzow and Petra, 2004 [9]), respectively, which commonly use the Fischer-Burmeister (FB) function to recast the mixed complementarity problem (MCP) as a constrained minimization problem and a nonlinear system of equations, respectively. They all remark that their algorithms may be improved if the FB function is replaced by other NCP functions. Accordingly, in this paper, we employ the generalized Fischer-Burmeister (GFB) where the 2-norm in the FB function is relaxed to a general p-norm (p > 1) for the two methods and investigate how much the improvement is by changing the parameter p as well as which method is influenced more when we do so, by the performance profiles of iterations and function evaluations for the two methods with different p on MCPLIB collection.

Original languageEnglish
Pages (from-to)667-683
Number of pages17
JournalJournal of Computational and Applied Mathematics
Volume234
Issue number3
DOIs
Publication statusPublished - 2010 Jun 1

Fingerprint

Mixed Complementarity Problem
Numerical Comparisons
NCP Function
Norm
Nonlinear Systems of Equations
Constrained Minimization
Function evaluation
Evaluation Function
Minimization Problem
Nonlinear systems
Iteration

Keywords

  • Convergence rate
  • MCP
  • Semismooth
  • The generalized FB function

ASJC Scopus subject areas

  • Computational Mathematics
  • Applied Mathematics

Cite this

Numerical comparisons of two effective methods for mixed complementarity problems. / Chen, Jein-Shan; Pan, Shaohua; Yang, Ching Yu.

In: Journal of Computational and Applied Mathematics, Vol. 234, No. 3, 01.06.2010, p. 667-683.

Research output: Contribution to journalArticle

@article{db1f7686d9084e97b177c3cdee18b05b,
title = "Numerical comparisons of two effective methods for mixed complementarity problems",
abstract = "Recently there have two different effective methods proposed by Kanzow et al. in (Kanzow, 2001 [8]) and (Kanzow and Petra, 2004 [9]), respectively, which commonly use the Fischer-Burmeister (FB) function to recast the mixed complementarity problem (MCP) as a constrained minimization problem and a nonlinear system of equations, respectively. They all remark that their algorithms may be improved if the FB function is replaced by other NCP functions. Accordingly, in this paper, we employ the generalized Fischer-Burmeister (GFB) where the 2-norm in the FB function is relaxed to a general p-norm (p > 1) for the two methods and investigate how much the improvement is by changing the parameter p as well as which method is influenced more when we do so, by the performance profiles of iterations and function evaluations for the two methods with different p on MCPLIB collection.",
keywords = "Convergence rate, MCP, Semismooth, The generalized FB function",
author = "Jein-Shan Chen and Shaohua Pan and Yang, {Ching Yu}",
year = "2010",
month = "6",
day = "1",
doi = "10.1016/j.cam.2010.01.004",
language = "English",
volume = "234",
pages = "667--683",
journal = "Journal of Computational and Applied Mathematics",
issn = "0377-0427",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Numerical comparisons of two effective methods for mixed complementarity problems

AU - Chen, Jein-Shan

AU - Pan, Shaohua

AU - Yang, Ching Yu

PY - 2010/6/1

Y1 - 2010/6/1

N2 - Recently there have two different effective methods proposed by Kanzow et al. in (Kanzow, 2001 [8]) and (Kanzow and Petra, 2004 [9]), respectively, which commonly use the Fischer-Burmeister (FB) function to recast the mixed complementarity problem (MCP) as a constrained minimization problem and a nonlinear system of equations, respectively. They all remark that their algorithms may be improved if the FB function is replaced by other NCP functions. Accordingly, in this paper, we employ the generalized Fischer-Burmeister (GFB) where the 2-norm in the FB function is relaxed to a general p-norm (p > 1) for the two methods and investigate how much the improvement is by changing the parameter p as well as which method is influenced more when we do so, by the performance profiles of iterations and function evaluations for the two methods with different p on MCPLIB collection.

AB - Recently there have two different effective methods proposed by Kanzow et al. in (Kanzow, 2001 [8]) and (Kanzow and Petra, 2004 [9]), respectively, which commonly use the Fischer-Burmeister (FB) function to recast the mixed complementarity problem (MCP) as a constrained minimization problem and a nonlinear system of equations, respectively. They all remark that their algorithms may be improved if the FB function is replaced by other NCP functions. Accordingly, in this paper, we employ the generalized Fischer-Burmeister (GFB) where the 2-norm in the FB function is relaxed to a general p-norm (p > 1) for the two methods and investigate how much the improvement is by changing the parameter p as well as which method is influenced more when we do so, by the performance profiles of iterations and function evaluations for the two methods with different p on MCPLIB collection.

KW - Convergence rate

KW - MCP

KW - Semismooth

KW - The generalized FB function

UR - http://www.scopus.com/inward/record.url?scp=77949486393&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949486393&partnerID=8YFLogxK

U2 - 10.1016/j.cam.2010.01.004

DO - 10.1016/j.cam.2010.01.004

M3 - Article

AN - SCOPUS:77949486393

VL - 234

SP - 667

EP - 683

JO - Journal of Computational and Applied Mathematics

JF - Journal of Computational and Applied Mathematics

SN - 0377-0427

IS - 3

ER -