A Study on Portfolio Selection Based on Fuzzy Linear Programming

AKBAŞ S., ERBAY DALKILIÇ T., Aksoy T. G.

INTERNATIONAL JOURNAL OF UNCERTAINTY FUZZINESS AND KNOWLEDGE-BASED SYSTEMS, cilt.30, sa.02, ss.211-230, 2022 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 30 Sayı: 02
  • Basım Tarihi: 2022
  • Doi Numarası: 10.1142/s021848852250009x
  • Dergi Adı: INTERNATIONAL JOURNAL OF UNCERTAINTY FUZZINESS AND KNOWLEDGE-BASED SYSTEMS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Business Source Elite, Business Source Premier, Computer & Applied Sciences, zbMATH
  • Sayfa Sayıları: ss.211-230
  • Anahtar Kelimeler: Fuzzy logic, linear programming, portfolio optimization, ISE-30, S&P500, OPTIMIZATION MODEL, MULTIPLE CRITERIA, INVESTMENT, SYSTEM, RISK
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

Portfolio management is the allocation of funds in the hands of the investor to ensure minimum risk and maximum profitability among the existing securities. In the finance sector related to portfolio management, many approaches, theories, and models have been developed. In this study, a new model is proposed. The basic objective of this model is to minimize the risk of portfolio while maximizing expected return of the portfolio. The proposed model is based on the Mean Absolute Deviation Model (MAD) proposed by Konno-Yamazaki. Considering the uncertainty of expected returns in the Mean Absolute Deviation Model, the problem was remodeled with fuzzy logic approach. In the application section, the proposed model has been applied to two different indexes, namely Istanbul Stock Exchange (ISE-30) and Standard & Poor's 500 (S&P 500). In application, 64-month return movements of stocks traded in both indexes between 01.01.2016-30.04.2021 were used. In order to examine the effectiveness of the model, firstly, monthly return data for both indexes were modeled with the Mean-Absolute Deviation method and the Mean Variance method. Then, the same data were modeled with the proposed method. In the conclusion section of the study, the results obtained from each method were compared.