欢迎来到天天文库
浏览记录
ID:32074744
大小:5.83 MB
页数:50页
时间:2019-01-31
《投资组合优化模型的文化算法分析 (1)》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、AbstractPortfolioisanimportantpartofmodemfinancewhichprovidesawaytoallocatethecapitaltoseveralriskassetsinordertominimizetheriskormaximizetheprofit.Atpresent,optimizationmethodsandadvancedcomputationaltechnologyare’usedtosolvetheportfoliooptimizationmod
2、els,especiallytheapplicationofintelligentalgorithms.CulturalalgorithmisarecentlyemergingbionicmteIllgentalgorithmthatprovidesauniversalframeworkofthedouble-layerevolution,whileovercomangthedeftcienciesofthetraditionalintelligentalgorithmfortherecessivee
3、xpressionandacce88ottneexperienceandknowledge.Withthehelpoftheuniquecomputationalframeworkofculturalalgorithmandthegoodparallel,thisthesiseffectivelyimprovesthealgorithmbyusingtheideaofintegrationandappliestheimprovedhybridalgorithmforsolvingthefunction
4、optimizationandportfolio0pt蛐1zatlonmodels.Themainworkofthisthesisisasfollows:1.ThisthesismakestheparticleswarmoptimizationembedintoculturalalgorithmasitspopulatlonsDaceandbeliefspace,andproposesanewdynamicparticleswarmoptlmlzatmnbasedonculnlraIalgorithm
5、.AsetofdynamiccommunicationoperationisdesignedtoachievetheinteractionbetweenthetwosDacesinordertoimproveandovercomethedefectthattheparticleswarmoptimizatlon1seasllYtrappedintothelocaloptimum.ExperimentalresultsofseveralBenchmarkfunctionsshowthatthisalgo
6、rithmnotonlyhasahighprecisionbutalsohasastrongrobustness,anditisaneffectiveglobaloptimizationalgorithm.Portfoliodecisionfacingagreatcomplicatedcombinatorialoptimizationproblem,andisdealofdataoftherealsecuritymarketisaNP.hardproblem,whichisdifficulttobes
7、olvedbytraditionalalgorithm.Thisthesisintroducesanewdynamicparticleswarmopti叫zatlonbased0nculturalalgorithmforsolvingthemean·VaRmodel,andusesthepenaltyfunctionapproachtotheinequalityconstraintsinthemodel.Anempiricalanalysisresultsshowthattheproblemofpor
8、tfoliooDtimizationcanbesolvedmorereasonablyandefficientlybythisalgorithm.2.Basedextemalelitearchiveandcrowdingentropystrategy,thisthesisgwesanewmult卜objectiveDarticleswarrnoptimizationbasedonculturalalgorithm.Firstly,itistestedon
此文档下载收益归作者所有