ارزیابی مقایسه‌ای رویکرد مارکویتز با یک روش ترکیبی به منظور تشکیل پرتفوی بهینه با کاربرد یادگیری عمیق DNN و الگوریتم جستجوی گرانشی

نوع مقاله : علمی - پژوهشی

نویسندگان

1 استادیار بخش اقتصاد، دانشگاه یزد، یزد، ایران.

2 دانشجوی دکتری مهندسی مالی، دانشگاه یزد، یزد، ایران.

3 دانشیار بخش مالی و حسابداری، دانشگاه یزد، یزد، ایران

چکیده

هدف پژوهش حاضر، مقایسه عملکرد روشی ترکیبی نوآورانه با عملکرد بهینه‌سازی سبد سهام به روش معمول مارکوییتز است. بدین منظور ابتدا با استفاده از یک شبکه یادگیری عمیق DNN و متغیرهای تکنیکی سهام برای بازه 02/04/1397 تا 02/06/1397 به پیش‌بینی قیمت آتی سهام پرداخته شد؛ سپس بر اساس قیمت‌های آتی سهام، بازده و ریسک سهام محاسبه و سود پرتفو با قید ریسک و با روش الگوریتم گرانشی از طریق نرم افزار متلب حداکثر شد. این عمل به ایجاد پرتفوی‌های ریسک­گریز تا ریسک‌پذیر روی مرز کارای پارتو منجر می‌شود. پس از آن بازدهی آتی پرتفوها برای دو ماه آینده محاسبه و فرایند ذکر­شده برای 30 هفته به شکل پنجره غلطان و با گام‌های یک‌هفته‌ای تکرار شد. این نتایج با نتایج حاصل از روش عادی مارکوییتز و با بهینه­سازی از طریق الگوریتم جست‌وجوی گرانشی مبتنی بر شاخص‌های تکنیکی برای 30 دوره مقایسه شد. نتایج نشان داد که روش مبتنی بر پیش‌بینی قیمت سهام با استفاده از شاخص‌های تکنیکی و همچنین روش مارکویتز تنها در پرتفوی ریسک­گریز عملکرد بهتری نبست به میانگین شاخص بازار ارائه می­دهد. 

کلیدواژه‌ها

عنوان مقاله [English]

Comparative Evaluation of Markowitz Approach with a New Hybrid Method to Create an Optimal Portfolio Using Deep DNN Learning Method and Gravitational Search Algorithm.

نویسندگان [English]

  • Mohammad Hassan Zare 1
  • Moslem Nilchi 2
  • Daryush Fareed 3
1 Assistant Prof, Department of Economics, Yazd Universtity, Yazd, Iran.
2 Ph.D. Candidate in Financial Engineering, Yazd Universtity, Yazd, Iran.
3 Associate Prof, Department of Finance and Acconting, Yazd Universtity, Yazd, Iran
چکیده [English]

The aim of this study is to compare the New Hybrid Method with the usual Markowitz method in creating an optimal portfolio.  To this end, at the first stage, the future stock prices were predicted using a deep DNN learning method and stock technical variables for the period 1397/4/2 to 1397/6/2. Then, based on future stock prices, stock return and risk were calculated and, by using Gravitational Algorithm, portfolio profits were maximized. This results in creating low risk to high risk portfolios on the Pareto efficient frontier. After that, the future return of portfolios was calculated for the next two months, and the process was repeated for 30 weeks in the form of weekly Rolling Window. These results were compared with the results of usual Markovitz method for 30 periods. The results indicated that both Markowitz and New Hybrid methods showed only better performance in predicting stock prices of risk averse portfolios than average market index.  

کلیدواژه‌ها [English]

  • Hybrid Approach
  • Stock Portfolio
  • Gravitational Research Approach
  • Deep Learning

مراجع

  1. - Abbasi, Ebrahim; Abvali, Mehdi; Sarbazi, Mehdi (2012). Choosing the Optimal Stock Portfolio Using NSGA-II Genetic Algorithm. Journal of Financial Engineering and Management Securities (10) (in Persian).
  2. - Abzari, mehdi; ketabi, Saeedeh; Abbasi, Abbas (2005). Optimizing Investment Portfolio Using Linear Programing Methods and Providing a Functional Model. Journal of Social and Humanities Sciences of Shiraz University, 22(2) (in Persian).
  3. - Afsar, Amir; Halil, Fatemeh (2017). Portfolio Optimization with Hybrid Approach of Technical Analysis and Data Mining Methods. Journal on New Research in Decision Making, 2(2) (in Persian).
  4. - Alizadeh Noodehi, Elaheh; Mahfoozi, Gholamreza; Vasiresh, Atieh (2015). Comparison of Efficiency and Risk in Technical Method with Balanced and Reinvestment and Purchase and Maintenance Strategy in Tehran Stock Exchange. Journal of Investment Knowledge (16) (in Persian).
  5. - Bayat, Ali; Asadi, Lida (2017). Optimization of Stock Portfolio: Utility of Birds Algorithm and Markoitz Model. Financial Engineering and Managing Securities 8(32) (in Persian).
  6. - Berger, T., Berger, T., Fieberg, C., & Fieberg, C. (2016). On portfolio optimization: Forecasting asset covariances and variances based on multi-scale risk models. The Journal of Risk Finance, 17(3), 295-309.
  7. - Cheraghi, Babak (2000), Predicting Stock price Behavior in The Framework of Technical Analysis Model: Case Study of Tehran Stock Exchange. Thesis of Master in Economics, Faculty of Economics, Tehran University.
  8. - Fadaee nejad, Mohammad Esmaeel; Sadeghi, Mohsen (2006). Assessing the Usefulness of Momentum and Reverse Strategies. Journal of Management Perspective 17 (in Persian).
  9. - Gudarzi M., Yakideh K., Mahfuzi G. (2016), Portfolio optimization by combining data envelopment analysis and decision-making Hurwicz Method. Journal of Modern Researches in Decision Making 1(4).
  10. - Huang, X. (2008). Portfolio selection with a new definition of risk. European Journal of Operational Research, 186(1), 351-357.
  11. - Jahankhani, Ali; Parsaeian, Ali (2015). Financial management, Tehran: Samt Publication.
  12. - Kalayci, C. B., Polat, O., & Akbay, M. A. (2020). An efficient hybrid metaheuristic algorithm for cardinality constrained portfolio optimization. Swarm and Evolutionary Computation 54.
  13. - Keyani Herchgani, Maedeh; Nabavi Chashmi, Seyed Ali; Memarian, Erfan (2014). Portfolio Optimization based on Minimum Level of Total Risk Acceptance and Its Components Using Genetic Algorithm Method. Journal of Investment Knowledge 3(11) (in Persian).
  14. - Khalili Osbouee, Saber (2013), Assessing of Strategic Financial Performance of Companies in Tehran Stock Exchange by Employing Multi- Criteria Decision making Texts in Fuzzy Environment. Journal of Monetary and Banking Management Development 1(1) (in Persian).
  15. - Khanjarpanah, Hossein; doroush, Davood; Shavalpour, Saeed; Jabbarzadeh, Armin (2018). Application of Technical Methods for Predicting Stock Prices: The Approach of Non-linear Probability Models and Artificial Neural Networks. Journal of Financial management Strategy 6(3) (in Persian).
  16. - Loreggia, A., Malitsky, Y., Samulowitz, H., & Saraswat, V. (2016, February). Deep learning for algorithm portfolios. In Thirtieth AAAI Conference on Artificial Intelligence.
  17. - Macedo, L. L., Godinho, P., & Alves, M. J. (2017). Mean-semivariance portfolio optimization with multiobjective evolutionary algorithms and technical analysis rules. Expert Systems with Applications, 79, 33-43.
  18. - Manafi, Ebrahim (2015). Adjustment of Return and Its Impact on Optimum Bound Portfolio Performance in Mean-Half Variance Model. Thesis of Master in Economics, Faculty of Economics, Tehran University.
  19. - Markowitz, H. M. (1976). Markowitz revisited. Financial Analysis Journal, 32(5), 47-52.
  20. - Mirzaee, Hamid Reza; Khodamipour, Ahmad; Pourheidari, Omid (2016). Application of Multi Objective Genetic Algorithm for Optimization of Stock Portfolios Using Technical Indicatores. Financial Engineering and Securities Management 7(29).
  21. - Raee, Reza (2001). Neural Networks: A New Approach in Making Decisions in Management. Journal of Humanities 19.
  22. - Raee, Reza; Saeedi, Ali (2009). Fundamentals of financial Engineering and Risk Management, Tehran: Samt Publication.
  23. - Raee, Reza; Talangi, Ahmad (2016). Advanced Investment Management, Tehran: Samt Publication.
  24. - Rashedi, E., Nezamabadi-Pour, H. and Saryazdi, S., (2009). "GSA: a gravitational search algorithm". Information sciences, 112(10), pp.9909-9943.
  25. - Razmi, Jaafar; Jouly, Fariborz; Tavakoli Moghadam, Reza; Abbaslou, Amir Abbas (2009). Evaluation of Efficiency of Technical Analysis Methods in Tehran Stock Exchange. Journal of Industrial Engineering 43(1) (in Persian).
  26. - Saberi, Maryam; Darabi, Roya; Hamidian, Mohsen (2019). Optimal Portfolios in Speculation Bubble Space Based on Subjective Accounting. Journal of Knowledge Investment 8(30) (in Persian).
  27. - Shynkevich, A. (2016). Predictability in bond returns using technical trading rules. Journal of Banking & Finance, 70, 55-69.
  28. - Silva, A., Neves, R., & Horta, N. (2015). A hybrid approach to portfolio composition based on fundamental and technical indicators. Expert Systems with Applications, 42(4), 2036-2048.
  29. - Tehrani, Reza; Esmaeeli, Mohammad (2012). The Effect of Using Important Indicators of Technical Analysis on Short -term Returns of Investment in Tehran Stock Market. Financial Knowledge of Stock market Analysis (Financial Studies) 5(13) (in Persian).

فایل ها

هم رسانی

ارجاع به این مقاله

آمار