Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa Istanbul
We introduce a novel portfolio optimization framework—Distributionally Robust Multivariate Stochastic Cone Order (DR-MSCO)—which integrates partial orders on random vectors with Wasserstein-metric ambiguity sets and adaptive cone structures to model multivariate investor preferences under distributi...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/15/2473 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849770880811401216 |
|---|---|
| author | Larissa Margerata Batrancea Mehmet Ali Balcı Ömer Akgüller Lucian Gaban |
| author_facet | Larissa Margerata Batrancea Mehmet Ali Balcı Ömer Akgüller Lucian Gaban |
| author_sort | Larissa Margerata Batrancea |
| collection | DOAJ |
| description | We introduce a novel portfolio optimization framework—Distributionally Robust Multivariate Stochastic Cone Order (DR-MSCO)—which integrates partial orders on random vectors with Wasserstein-metric ambiguity sets and adaptive cone structures to model multivariate investor preferences under distributional uncertainty. Grounded in measure theory and convex analysis, DR-MSCO employs data-driven cone selection calibrated to market regimes, along with coherent tail-risk operators that generalize Conditional Value-at-Risk to the multivariate setting. We derive a tractable second-order cone programming reformulation and demonstrate statistical consistency under empirical ambiguity sets. Empirically, we apply DR-MSCO to 23 Borsa Istanbul equities from 2021–2024, using a rolling estimation window and realistic transaction costs. Compared to classical mean–variance and standard distributionally robust benchmarks, DR-MSCO achieves higher overall and crisis-period Sharpe ratios (2.18 vs. 2.09 full sample; 0.95 vs. 0.69 during crises), reduces maximum drawdown by 10%, and yields endogenous diversification without exogenous constraints. Our results underscore the practical benefits of combining multivariate preference modeling with distributional robustness, offering institutional investors a tractable tool for resilient portfolio construction in volatile emerging markets. |
| format | Article |
| id | doaj-art-ff81fba90b49493f8d88539f73b62c93 |
| institution | DOAJ |
| issn | 2227-7390 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-ff81fba90b49493f8d88539f73b62c932025-08-20T03:02:49ZengMDPI AGMathematics2227-73902025-07-011315247310.3390/math13152473Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa IstanbulLarissa Margerata Batrancea0Mehmet Ali Balcı1Ömer Akgüller2Lucian Gaban3Department of Business, Babeş-Bolyai University, 7 Horea Street, 400174 Cluj-Napoca, RomaniaDepartment of Mathematics, Faculty of Science, Mugla Sitki Kocman University, Muğla 48000, TurkeyDepartment of Mathematics, Faculty of Science, Mugla Sitki Kocman University, Muğla 48000, TurkeyFaculty of Economics, “1 Decembrie 1918” University of Alba Iulia, 510009 Alba Iulia, RomaniaWe introduce a novel portfolio optimization framework—Distributionally Robust Multivariate Stochastic Cone Order (DR-MSCO)—which integrates partial orders on random vectors with Wasserstein-metric ambiguity sets and adaptive cone structures to model multivariate investor preferences under distributional uncertainty. Grounded in measure theory and convex analysis, DR-MSCO employs data-driven cone selection calibrated to market regimes, along with coherent tail-risk operators that generalize Conditional Value-at-Risk to the multivariate setting. We derive a tractable second-order cone programming reformulation and demonstrate statistical consistency under empirical ambiguity sets. Empirically, we apply DR-MSCO to 23 Borsa Istanbul equities from 2021–2024, using a rolling estimation window and realistic transaction costs. Compared to classical mean–variance and standard distributionally robust benchmarks, DR-MSCO achieves higher overall and crisis-period Sharpe ratios (2.18 vs. 2.09 full sample; 0.95 vs. 0.69 during crises), reduces maximum drawdown by 10%, and yields endogenous diversification without exogenous constraints. Our results underscore the practical benefits of combining multivariate preference modeling with distributional robustness, offering institutional investors a tractable tool for resilient portfolio construction in volatile emerging markets.https://www.mdpi.com/2227-7390/13/15/2473distributionally robust optimizationmultivariate stochastic cone orderWasserstein ambiguityportfolio optimization |
| spellingShingle | Larissa Margerata Batrancea Mehmet Ali Balcı Ömer Akgüller Lucian Gaban Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa Istanbul Mathematics distributionally robust optimization multivariate stochastic cone order Wasserstein ambiguity portfolio optimization |
| title | Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa Istanbul |
| title_full | Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa Istanbul |
| title_fullStr | Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa Istanbul |
| title_full_unstemmed | Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa Istanbul |
| title_short | Distributionally Robust Multivariate Stochastic Cone Order Portfolio Optimization: Theory and Evidence from Borsa Istanbul |
| title_sort | distributionally robust multivariate stochastic cone order portfolio optimization theory and evidence from borsa istanbul |
| topic | distributionally robust optimization multivariate stochastic cone order Wasserstein ambiguity portfolio optimization |
| url | https://www.mdpi.com/2227-7390/13/15/2473 |
| work_keys_str_mv | AT larissamargeratabatrancea distributionallyrobustmultivariatestochasticconeorderportfoliooptimizationtheoryandevidencefromborsaistanbul AT mehmetalibalcı distributionallyrobustmultivariatestochasticconeorderportfoliooptimizationtheoryandevidencefromborsaistanbul AT omerakguller distributionallyrobustmultivariatestochasticconeorderportfoliooptimizationtheoryandevidencefromborsaistanbul AT luciangaban distributionallyrobustmultivariatestochasticconeorderportfoliooptimizationtheoryandevidencefromborsaistanbul |