<italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio Optimization
<p>Portfolio optimization is a classical and important problem in the field of asset management, which aims to achieve a trade-off between profit and risk. Previous portfolio optimization models use traditional risk measurements such as variance, which symmetrically delineate both positive and...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2023-09-01
|
| Series: | Complex System Modeling and Simulation |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.23919/CSMS.2023.0012 |
| _version_ | 1827091281850925056 |
|---|---|
| author | Feng Wang Zilu Huang Shuwen Wang |
| author_facet | Feng Wang Zilu Huang Shuwen Wang |
| author_sort | Feng Wang |
| collection | DOAJ |
| description | <p>Portfolio optimization is a classical and important problem in the field of asset management, which aims to achieve a trade-off between profit and risk. Previous portfolio optimization models use traditional risk measurements such as variance, which symmetrically delineate both positive and negative sides and are not practical and stable. In this paper, a new model with cardinality constraints is first proposed, in which the idiosyncratic volatility factor is used to replace traditional risk measurements and can capture the risks of the portfolio in a more accurate way. The new model has practical constraints which involve the sparsity and irregularity of variables and make it challenging to be solved by traditional Multi-Objective Evolutionary Algorithms (MOEAs). To solve the model, a Learning-Guided Evolutionary Algorithm based on <inline-formula id="Z-20230619135908"><math id="mathml_Z-20230619135908" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula> indicator (<inline-formula id="M4"><math id="mathml_M4" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula>LGEA) is developed. In <inline-formula id="M5"><math id="mathml_M5" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula>LGEA, the <inline-formula id="M6"><math id="mathml_M6" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula> indicator is incorporated into the initialization and genetic operators to guarantee the sparsity of solutions and can help improve the convergence of the algorithm. And a new constraint-handling method based on <inline-formula id="M7"><math id="mathml_M7" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula> indicator is also adopted to ensure the feasibility of solutions. The experimental results on five portfolio trading datasets including up to 1226 assets show that <inline-formula id="M8"><math id="mathml_M8" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula>LGEA outperforms some state-of-the-art MOEAs in most cases.</p> |
| first_indexed | 2024-03-11T19:04:58Z |
| format | Article |
| id | doaj.art-0faf4fd8df5b4153b6aa88c53d4560bf |
| institution | Directory Open Access Journal |
| issn | 2096-9929 |
| language | English |
| last_indexed | 2025-03-20T05:53:49Z |
| publishDate | 2023-09-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Complex System Modeling and Simulation |
| spelling | doaj.art-0faf4fd8df5b4153b6aa88c53d4560bf2024-10-03T01:16:44ZengTsinghua University PressComplex System Modeling and Simulation2096-99292023-09-013319120110.23919/CSMS.2023.0012<italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio OptimizationFeng Wang0Zilu Huang1Shuwen Wang2School of Computer Science, Wuhan University, Wuhan 430072, ChinaSchool of Computer Science, Wuhan University, Wuhan 430072, ChinaSloan School of Management, Massachusetts Institute of Technology, Cambridge, MA 02139, USA<p>Portfolio optimization is a classical and important problem in the field of asset management, which aims to achieve a trade-off between profit and risk. Previous portfolio optimization models use traditional risk measurements such as variance, which symmetrically delineate both positive and negative sides and are not practical and stable. In this paper, a new model with cardinality constraints is first proposed, in which the idiosyncratic volatility factor is used to replace traditional risk measurements and can capture the risks of the portfolio in a more accurate way. The new model has practical constraints which involve the sparsity and irregularity of variables and make it challenging to be solved by traditional Multi-Objective Evolutionary Algorithms (MOEAs). To solve the model, a Learning-Guided Evolutionary Algorithm based on <inline-formula id="Z-20230619135908"><math id="mathml_Z-20230619135908" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula> indicator (<inline-formula id="M4"><math id="mathml_M4" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula>LGEA) is developed. In <inline-formula id="M5"><math id="mathml_M5" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula>LGEA, the <inline-formula id="M6"><math id="mathml_M6" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula> indicator is incorporated into the initialization and genetic operators to guarantee the sparsity of solutions and can help improve the convergence of the algorithm. And a new constraint-handling method based on <inline-formula id="M7"><math id="mathml_M7" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula> indicator is also adopted to ensure the feasibility of solutions. The experimental results on five portfolio trading datasets including up to 1226 assets show that <inline-formula id="M8"><math id="mathml_M8" display="inline" overflow="scroll"><msub><mi>I</mi><mrow class="MJX-TeXAtom-ORD"><mi>ϵ</mi><mo>+</mo></mrow></msub></math></inline-formula>LGEA outperforms some state-of-the-art MOEAs in most cases.</p>https://www.sciopen.com/article/10.23919/CSMS.2023.0012portfolio optimizationevolutionary algorithmsparse solution spaceindicator-based evolutionary algorithm (ea) |
| spellingShingle | Feng Wang Zilu Huang Shuwen Wang <italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio Optimization Complex System Modeling and Simulation portfolio optimization evolutionary algorithm sparse solution space indicator-based evolutionary algorithm (ea) |
| title | <italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio Optimization |
| title_full | <italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio Optimization |
| title_fullStr | <italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio Optimization |
| title_full_unstemmed | <italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio Optimization |
| title_short | <italic>I</italic><sub><italic>ϵ</italic>+</sub>LGEA: A Learning-Guided Evolutionary Algorithm Based on <italic>I</italic><sub><italic>ϵ</italic>+</sub> Indicator for Portfolio Optimization |
| title_sort | italic i italic sub italic ϵ italic sub lgea a learning guided evolutionary algorithm based on italic i italic sub italic ϵ italic sub indicator for portfolio optimization |
| topic | portfolio optimization evolutionary algorithm sparse solution space indicator-based evolutionary algorithm (ea) |
| url | https://www.sciopen.com/article/10.23919/CSMS.2023.0012 |
| work_keys_str_mv | AT fengwang italiciitalicsubitaliceitalicsublgeaalearningguidedevolutionaryalgorithmbasedonitaliciitalicsubitaliceitalicsubindicatorforportfoliooptimization AT ziluhuang italiciitalicsubitaliceitalicsublgeaalearningguidedevolutionaryalgorithmbasedonitaliciitalicsubitaliceitalicsubindicatorforportfoliooptimization AT shuwenwang italiciitalicsubitaliceitalicsublgeaalearningguidedevolutionaryalgorithmbasedonitaliciitalicsubitaliceitalicsubindicatorforportfoliooptimization |