Back to Search
Start Over
A penalty decomposition algorithm for the extended mean–variance–CVaR portfolio optimization problem.
- Source :
-
Discrete Mathematics, Algorithms & Applications . Apr2024, Vol. 16 Issue 3, p1-16. 16p. - Publication Year :
- 2024
-
Abstract
- In this paper, we study mean–variance–Conditional Value-at-Risk (CVaR) portfolio optimization problem with short selling, cardinality constraint and transaction costs. To tackle its mixed-integer quadratic optimization model for large number of scenarios, we take advantage of the penalty decomposition method (PDM). It needs solving a quadratic optimization problem and a mixed-integer linear program at each iteration, where the later one has explicit optimal solution. The convergence of PDM to a partial minimum of original problem is proved. Finally, numerical experiments using the S&P index for 2020 are conducted to evaluate efficiency of the proposed algorithm in terms of return, variance and CVaR gaps and CPU times. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 17938309
- Volume :
- 16
- Issue :
- 3
- Database :
- Academic Search Index
- Journal :
- Discrete Mathematics, Algorithms & Applications
- Publication Type :
- Academic Journal
- Accession number :
- 175283980
- Full Text :
- https://doi.org/10.1142/S1793830923500210