A bi-level optimization heuristic for solving portfolio selection problem
Author(s):
Article Type:
Research/Original Article (دارای رتبه معتبر)
Abstract:
This research addresses the challenge of optimizing the selection of a stock portfolio within a scenario involving two distinct decision makers, each driven by their unique objective functions and constraints. Importantly, the choices made by one decision-maker impact the decisions of the other. The primary focus is a dual-level model: at the first level (leader), investment management firms' decisions are expressed as they strive to partake in trading profits through portfolio management. At the second level (follower), active investors in the capital market are examined, pursuing goals of maximizing returns and minimizing investment risk. A notable innovation lies in incorporating social criteria and implicit investor preferences into the considerations of portfolio management firms. Given that solving two-level optimization problems with variable variability at the follower level is recognized as a complex polynomial challenge, this research introduces an inventive algorithm based on exhaustive enumeration to tackle the proposed model. Numerical outcomes stemming from portfolio optimization, utilizing data from the Tehran Stock Exchange market and Iran Farabourse market. Furthermore, numerical analyses underscore the favorable performance of the proposed model and algorithm in addressing issues related to stock portfolio optimization. Thus, this model holds promise as a management tool for similar challenges.
Keywords:
Language:
English
Published:
International Journal of Finance and Managerial Accounting, Volume:11 Issue: 41, Spring 2026
Pages:
123 to 138
https://www.magiran.com/p2807187
سامانه نویسندگان
مقالات دیگری از این نویسنده (گان)
-
A Hybrid Decision-Making Model for Optimal Portfolio Selection under Interval Uncertainty
Mohsen Zahmati Iraj, *
Iranian Journal of Accounting, Auditing and Finance, Autumn 2024 -
Developing a Stock Market Prediction Model by Deep Learning Algorithms
Omid Boroumand, *
Journal of Information Technology Management, Summer 2024