View in EDS PDF Full Text

Varlık sayısı kısıtlı portföy optimizasyon probleminin çözümünde yazılım programlarının performans karşılaştırması.

Saved in:
Bibliographic Details
Title: Varlık sayısı kısıtlı portföy optimizasyon probleminin çözümünde yazılım programlarının performans karşılaştırması.
Alternate Title: Performance comparison of software programs in solving the cardinality constrained portfolio optimization problem.
Authors: Deliktaş, Derya1 derya.delktas@dpu.edu.tr, Tutumlu, Büşra1 busra.tutumlu@dpu.edu.tr, Üstün, Özden2 ozden.ustun@aksaray.edu.tr
Source: Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi,. Mar2026, Vol. 41 Issue 1, p595-608. 14p.
Subjects: Optimizers (Computer software), Portfolio management (Investments), Financial management, Mathematics software, Computer performance, Benchmark problems (Computer science), Computer software
Abstract (English): Cardinality constrained portfolio optimization is one of the important problems in financial management, where the number of assets in the portfolio is limited. Unlike classical portfolio optimization, this problem is more complex in terms of computation since it also involves asset selection. In the literature, optimization software packages are widely used to solve this problem using exact solution methods. Commonly used software packages are Lingo, Gams/Dicopt, and Gurobi. In this study, the performances of the software programs in terms of solution quality, computational efficiency, and practicality in application were analyzed and compared by using these software programs in solving the cardinality-constrained portfolio optimization problem. The analyses were performed on data sets belonging to the Hang Seng, DAX 100, FTSE 100, S&P 100, and Nikkei 225 stock market indices, which are widely used in the literature. To make a correct and unbiased comparison, each software program was applied under the same conditions by maintaining consistency in parameters, restrictions, and calculation settings. Additionally, sensitivity analysis was conducted based on the number of assets. As a result of the comparison, it was seen that Gurobi is more successful than other solvers in terms of both computation time and better performance metric values. In addition, Gurobi's good performance and higher adaptability to large data sets are due to its advanced solver algorithms. [ABSTRACT FROM AUTHOR]
Abstract (Turkish): Varlık sayısı kısıtlı portföy optimizasyonu, portföydeki varlık sayısının sınırlı olduğu finansal yönetimde önemli problemlerden biridir. Klasik portföy optimizasyonu aksine, bu problemde ayrıca varlık seçimi söz konusu olması nedeniyle hesaplama açısından daha karmaşıktır. Literatürde bu problemi kesin çözüm yöntemleri kullanarak çözmek için yaygın olarak optimizasyon yazılım paket programları kullanılmaktadır. Genellikle kullanılan yazılım paket programları, Lingo, Gams/Dicopt ve Gurobi'dır. Bu çalışmada, varlık sayısı kısıtlı portföy optimizasyonu probleminin çözümünde bu yazılım programları kullanılarak çözüm kalitesi, hesaplama verimliliği ve uygulamada pratiklik açısından yazılım programlarının performansları analiz edilmiş ve karşılaştırılmıştır. Analizler, literatürde yaygın olarak kullanılan Hang Seng, DAX 100, FTSE 100, S&P 100 ve Nikkei 225 borsa endekslerine ait veri setleri üzerine gerçekleştirilmiştir. Doğru ve tarafsız bir karşılaştırma yapabilmek için her bir yazılım programı, parametreler, kısıtlamalar ve hesaplama ayarlarındaki tutarlılığı koruyarak aynı koşullar altında uygulanmıştır. Ayrıca varlık sayısına göre duyarlılık analizi yapılmıştır. Karşılaştırma sonucunda Gurobi'nin hem hesaplama süresi hem de daha iyi performans metriği değerlerine sahip olması açısından diğer çözücülerden daha başarılı olduğu görülmüştür. Buna ek olarak, Gurobi'nin iyi bir performans sergilemesi ve büyük veri setlerine karşı daha yüksek uyum yeteneğine sahip olması, gelişmiş çözücü algoritmalarına sahip olmasından kaynaklanmaktadır. [ABSTRACT FROM AUTHOR]
Copyright of Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, is the property of Gazi University, Faculty of Engineering & Architecture and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:Cardinality constrained portfolio optimization is one of the important problems in financial management, where the number of assets in the portfolio is limited. Unlike classical portfolio optimization, this problem is more complex in terms of computation since it also involves asset selection. In the literature, optimization software packages are widely used to solve this problem using exact solution methods. Commonly used software packages are Lingo, Gams/Dicopt, and Gurobi. In this study, the performances of the software programs in terms of solution quality, computational efficiency, and practicality in application were analyzed and compared by using these software programs in solving the cardinality-constrained portfolio optimization problem. The analyses were performed on data sets belonging to the Hang Seng, DAX 100, FTSE 100, S&P 100, and Nikkei 225 stock market indices, which are widely used in the literature. To make a correct and unbiased comparison, each software program was applied under the same conditions by maintaining consistency in parameters, restrictions, and calculation settings. Additionally, sensitivity analysis was conducted based on the number of assets. As a result of the comparison, it was seen that Gurobi is more successful than other solvers in terms of both computation time and better performance metric values. In addition, Gurobi's good performance and higher adaptability to large data sets are due to its advanced solver algorithms. [ABSTRACT FROM AUTHOR]
ISSN:13001884
DOI:10.17341/gazimmfd.1635003