A learning-guided multi-objective evolutionary algorithm for constrained portfolio optimization

Khin Lwin; Rong Qu; Graham  Kendall

doi:10.1016/j.asoc.2014.08.026

A learning-guided multi-objective evolutionary algorithm for constrained portfolio optimization

Khin Lwin, Rong Qu, Graham Kendall

Research output: Contribution to journal › Article

Abstract

Portfoliooptimizationinvolvestheoptimalassignmentoflimitedcapitaltodifferentavailablefinancialassetstoachieveareasonabletrade-offbetweenprofitandriskobjectives.Inthispaper,westudiedtheextendedMarkowitz’smean-varianceportfoliooptimizationmodel.Weconsideredthecardinality,quan-tity,pre-assignmentandroundlotconstraintsintheextendedmodel.Thesefourreal-worldconstraintslimitthenumberofassetsinaportfolio,restricttheminimumandmaximumproportionsofassetsheldintheportfolio,requiresomespecificassetstobeincludedintheportfolioandrequiretoinvesttheassetsinunitsofacertainsizerespectively.Anefficientlearning-guidedhybridmulti-objectiveevolutionaryalgo-rithmisproposedtosolvetheconstrainedportfoliooptimizationproblemintheextendedmean-varianceframework.Alearning-guidedsolutiongenerationstrategyisincorporatedintothemulti-objectiveopti-mizationprocesstopromotetheefficientconvergencebyguidingtheevolutionarysearchtowardsthepromisingregionsofthesearchspace.Theproposedalgorithmiscomparedagainstfourexistingstate-of-the-artmulti-objectiveevolutionaryalgorithms,namelyNon-dominatedSortingGeneticAlgorithm(NSGA-II),StrengthParetoEvolutionaryAlgorithm(SPEA-2),ParetoEnvelope-basedSelectionAlgorithm(PESA-II)andParetoArchivedEvolutionStrategy(PAES).ComputationalresultsarereportedforpubliclyavailableOR-librarydatasetsfromsevenmarketindicesinvolvingupto1318assets.Experimentalresultsontheconstrainedportfoliooptimizationproblemdemonstratethattheproposedalgorithmsignificantlyoutperformsthefourwell-knownmulti-objectiveevolutionaryalgorithmswithrespecttothequalityofobtainedefficientfrontierintheconductedexperiments

Original language	English
Pages (from-to)	757-772
Journal	Applied Soft Computing Journal
Volume	24
Early online date	27 Aug 2014
DOIs	https://doi.org/10.1016/j.asoc.2014.08.026
Publication status	Published - Nov 2014

Access to Document

10.1016/j.asoc.2014.08.026Licence: CC BY-NC-ND

Fingerprint Dive into the research topics of 'A learning-guided multi-objective evolutionary algorithm for constrained portfolio optimization'. Together they form a unique fingerprint.

Cite this

Lwin, K., Qu, R., & Kendall, G. (2014). A learning-guided multi-objective evolutionary algorithm for constrained portfolio optimization. Applied Soft Computing Journal, 24, 757-772. https://doi.org/10.1016/j.asoc.2014.08.026

@article{fe7f636acbc94d8a9c63e6fd9bfb89ea,

title = "A learning-guided multi-objective evolutionary algorithm for constrained portfolio optimization",

abstract = "Portfoliooptimizationinvolvestheoptimalassignmentoflimitedcapitaltodifferentavailablefinancialassetstoachieveareasonabletrade-offbetweenprofitandriskobjectives.Inthispaper,westudiedtheextendedMarkowitz{\textquoteright}smean-varianceportfoliooptimizationmodel.Weconsideredthecardinality,quan-tity,pre-assignmentandroundlotconstraintsintheextendedmodel.Thesefourreal-worldconstraintslimitthenumberofassetsinaportfolio,restricttheminimumandmaximumproportionsofassetsheldintheportfolio,requiresomespecificassetstobeincludedintheportfolioandrequiretoinvesttheassetsinunitsofacertainsizerespectively.Anefficientlearning-guidedhybridmulti-objectiveevolutionaryalgo-rithmisproposedtosolvetheconstrainedportfoliooptimizationproblemintheextendedmean-varianceframework.Alearning-guidedsolutiongenerationstrategyisincorporatedintothemulti-objectiveopti-mizationprocesstopromotetheefficientconvergencebyguidingtheevolutionarysearchtowardsthepromisingregionsofthesearchspace.Theproposedalgorithmiscomparedagainstfourexistingstate-of-the-artmulti-objectiveevolutionaryalgorithms,namelyNon-dominatedSortingGeneticAlgorithm(NSGA-II),StrengthParetoEvolutionaryAlgorithm(SPEA-2),ParetoEnvelope-basedSelectionAlgorithm(PESA-II)andParetoArchivedEvolutionStrategy(PAES).ComputationalresultsarereportedforpubliclyavailableOR-librarydatasetsfromsevenmarketindicesinvolvingupto1318assets.Experimentalresultsontheconstrainedportfoliooptimizationproblemdemonstratethattheproposedalgorithmsignificantlyoutperformsthefourwell-knownmulti-objectiveevolutionaryalgorithmswithrespecttothequalityofobtainedefficientfrontierintheconductedexperiments",

author = "Khin Lwin and Rong Qu and Graham Kendall",

year = "2014",

month = nov,

doi = "10.1016/j.asoc.2014.08.026",

language = "English",

volume = "24",

pages = "757--772",

journal = "Applied Soft Computing",

issn = "1568-4946",

publisher = "Elsevier",

}

TY - JOUR

T1 - A learning-guided multi-objective evolutionary algorithm for constrained portfolio optimization

AU - Lwin, Khin

AU - Qu, Rong

AU - Kendall, Graham

PY - 2014/11

Y1 - 2014/11

N2 - Portfoliooptimizationinvolvestheoptimalassignmentoflimitedcapitaltodifferentavailablefinancialassetstoachieveareasonabletrade-offbetweenprofitandriskobjectives.Inthispaper,westudiedtheextendedMarkowitz’smean-varianceportfoliooptimizationmodel.Weconsideredthecardinality,quan-tity,pre-assignmentandroundlotconstraintsintheextendedmodel.Thesefourreal-worldconstraintslimitthenumberofassetsinaportfolio,restricttheminimumandmaximumproportionsofassetsheldintheportfolio,requiresomespecificassetstobeincludedintheportfolioandrequiretoinvesttheassetsinunitsofacertainsizerespectively.Anefficientlearning-guidedhybridmulti-objectiveevolutionaryalgo-rithmisproposedtosolvetheconstrainedportfoliooptimizationproblemintheextendedmean-varianceframework.Alearning-guidedsolutiongenerationstrategyisincorporatedintothemulti-objectiveopti-mizationprocesstopromotetheefficientconvergencebyguidingtheevolutionarysearchtowardsthepromisingregionsofthesearchspace.Theproposedalgorithmiscomparedagainstfourexistingstate-of-the-artmulti-objectiveevolutionaryalgorithms,namelyNon-dominatedSortingGeneticAlgorithm(NSGA-II),StrengthParetoEvolutionaryAlgorithm(SPEA-2),ParetoEnvelope-basedSelectionAlgorithm(PESA-II)andParetoArchivedEvolutionStrategy(PAES).ComputationalresultsarereportedforpubliclyavailableOR-librarydatasetsfromsevenmarketindicesinvolvingupto1318assets.Experimentalresultsontheconstrainedportfoliooptimizationproblemdemonstratethattheproposedalgorithmsignificantlyoutperformsthefourwell-knownmulti-objectiveevolutionaryalgorithmswithrespecttothequalityofobtainedefficientfrontierintheconductedexperiments

AB - Portfoliooptimizationinvolvestheoptimalassignmentoflimitedcapitaltodifferentavailablefinancialassetstoachieveareasonabletrade-offbetweenprofitandriskobjectives.Inthispaper,westudiedtheextendedMarkowitz’smean-varianceportfoliooptimizationmodel.Weconsideredthecardinality,quan-tity,pre-assignmentandroundlotconstraintsintheextendedmodel.Thesefourreal-worldconstraintslimitthenumberofassetsinaportfolio,restricttheminimumandmaximumproportionsofassetsheldintheportfolio,requiresomespecificassetstobeincludedintheportfolioandrequiretoinvesttheassetsinunitsofacertainsizerespectively.Anefficientlearning-guidedhybridmulti-objectiveevolutionaryalgo-rithmisproposedtosolvetheconstrainedportfoliooptimizationproblemintheextendedmean-varianceframework.Alearning-guidedsolutiongenerationstrategyisincorporatedintothemulti-objectiveopti-mizationprocesstopromotetheefficientconvergencebyguidingtheevolutionarysearchtowardsthepromisingregionsofthesearchspace.Theproposedalgorithmiscomparedagainstfourexistingstate-of-the-artmulti-objectiveevolutionaryalgorithms,namelyNon-dominatedSortingGeneticAlgorithm(NSGA-II),StrengthParetoEvolutionaryAlgorithm(SPEA-2),ParetoEnvelope-basedSelectionAlgorithm(PESA-II)andParetoArchivedEvolutionStrategy(PAES).ComputationalresultsarereportedforpubliclyavailableOR-librarydatasetsfromsevenmarketindicesinvolvingupto1318assets.Experimentalresultsontheconstrainedportfoliooptimizationproblemdemonstratethattheproposedalgorithmsignificantlyoutperformsthefourwell-knownmulti-objectiveevolutionaryalgorithmswithrespecttothequalityofobtainedefficientfrontierintheconductedexperiments

U2 - 10.1016/j.asoc.2014.08.026

DO - 10.1016/j.asoc.2014.08.026

M3 - Article

VL - 24

SP - 757

EP - 772

JO - Applied Soft Computing

JF - Applied Soft Computing

SN - 1568-4946

ER -