Skip to main navigation Skip to search Skip to main content

A new evolutionary algorithm based on MOEA/D for portfolio optimization

Heng Zhang, Yaoyu Zhao, Feng Wang, Anran Zhang, Pengwei Yang, Xiaoliang Shen

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

Abstract

The portfolio optimization problem is a multi-objective problem which takes risk and return as optimization objectives. It is complicated in reality with many restrictions which results in an complex pareto front. MOEA/D is a popular multi-objective evolutionary algorithm framework with decomposition method, which has widely been used to solve multi-objective problems. In order to solve portfolio optimization problem with complex pareto front more effectively, we propose a new algorithm named MOEA/D-CP based on MOEA/D, which utilizes a new weight vector generation approach to generate a evenly distributed set of weight vectors. The experimental results show that the MOEA/D-CP performs much better than algorithm based on original MOEA/D.
Original languageEnglish
Title of host publicationProceedings of 2018 Tenth International Conference on Advanced Computational Intelligence (ICACI)
PublisherIEEE
Pages831-836
ISBN (Electronic)978-1-5386-4362-4
DOIs
Publication statusPublished - Mar 2018
Externally publishedYes
Event10th International Conference on Advanced Computational Intelligence, ICACI 2018 - Xiamen, Fujian, China
Duration: 29 Mar 201831 Mar 2018

Publication series

NameProceedings -10th International Conference on Advanced Computational Intelligence, ICACI

Conference

Conference10th International Conference on Advanced Computational Intelligence, ICACI 2018
PlaceChina
CityXiamen, Fujian
Period29/03/1831/03/18

Research Keywords

  • Complex pareto front
  • Portfolio optimization
  • Weight vector generation

Fingerprint

Dive into the research topics of 'A new evolutionary algorithm based on MOEA/D for portfolio optimization'. Together they form a unique fingerprint.

Cite this