ارائه الگوریتم ترکیبی یادگیری ماشین و ترکیب سنجههای ریسک و نظریه فازی در انتخاب سبد سرمایه گذاری
محورهای موضوعی : مهندسی مالیدانیال محمدی 1 , سید جعفر سجادی 2 , عمران محمدی 3 , نعیم شکری 4
1 - گروه مهندسی مالی، دانشکده مهندسی صنایع، دانشگاه علم و صنعت ایران، تهران، ایران.
2 - گروه مهندسی مالی، دانشکده مهندسی صنایع، دانشگاه علم و صنعت ایران، تهران، ایران.
3 - گروه مهندسی مالی، دانشکده مهندسی صنایع، دانشگاه علم و صنعت ایران، تهران، ایران.
4 - گروه توسعه و برنامه ریزی اقتصادی، دانشکده مدیریت و اقتصاد، دانشگاه تربیت مدرس، تهران، ایران.
کلید واژه: بورس اوراق بهادار تهران, یادگیری ماشین, بیز ساده, ارزش در معرض ریسک شرطی (CVaR), سبد سرمایه گذاری,
چکیده مقاله :
بازده و ریسک دو عامل مهم و اساسی برای تصمیمگیری در حوزه مالی میباشند. پژوهش حاضر جهت یافتن پرتفوی بهینه برای سرمایهگذاری از سهام بورسی انجام گرفته و یکیاز روشهاییکه در حال حاضر محبوبیت زیادی در بین تحلیلگران و پژوهش-گران این حوزه شکل گرفته، روشهای مبتنیبر هوش مصنوعی و در پی آن روشهایی با هدف کاهش سنجههای ریسک میباشد. هدف پژوهش حاضر تشکیل پرتفوی بااستفاده از روشهای یادگیری ماشین، سنجه ریسک و ترکیب آن با نظریه فازی است، که بازدهای بهتر از بازده میانگین بازار داشته باشد. خروجی هر روش وارد الگوریتم جنگل تصادفی شده و پیشبینی بهوسیله این الگوریتم صورت میگیرد و در مرحله آخر، خروجی پیشبینی برای تشکیل سبد سرمایه وارد مدل بهینهسازی ارزش در معرض ریسک و ارزش در معرض ریسک شرطی با رویکرد نظریه فازی میشوند. اطلاعات سهمها بهصورت روزانه و بازه زمانی آن از ابتدای سال 1394 تا اواسط سال 1398 میباشد. در پایان هرکدام از این روشها و مراحل با بازده واقعی بازار مقایسه گردید. بر اساس نتایج بدست آمده سنجهریسک CVAR قابلیت بهتری را نسبتبه سنجه ریسک VAR داشته است، همچنین الگوریتم جنگل تصادفی در بین الگوریتمهای یادگیری ماشین استفاده شده، نتایج بهتری را در انتخاب سبد سرمایهگذاری رقم زده است.
The current research was conducted to find the optimal portfolio for investing in stock exchange stocks, and one of the methods that is currently very popular among analysts and researchers in this field is methods based on artificial intelligence, followed by methods aimed at reducing risk metrics. The aim of the current research is to form a portfolio using machine learning methods, risk measurement and its combination with fuzzy theory, which has a better return than the average return of the market. The output of each method is entered into the random forest algorithm and prediction is made by this algorithm, and in the last step, the prediction output is entered into the value-at-risk and value-at-risk optimization model with the fuzzy theory approach to form the capital portfolio. Shares information is daily and its time period is from the beginning of 2014 to the middle of 2018. At the end of each of these methods and steps, it was compared with the real return of the market. the CVAR risk measure has a better ability than the VAR risk measure, and the random forest algorithm among the used machine learning algorithms has achieved better results in choosing the investment portfolio.
_|1) Abzari, M. (2005) Optimizing the investment portfolio using linear programming methods and providing a practical model. Journal of social and human sciences of Shiraz University.
2) Alborzi, M. (2001) Familiar with artificial neural networks. Tehran: Institute of Scientific Publications.
3) Biglari, B. (2010) Comparison of stock selection models for portfolio formation in terms of expected return, actual future return and their risk in Tehran Stock Exchange. The fifth financial system development conference in Iran.
4) Beigi, A. (2010) Optimizing the stock portfolio using the method of cumulative movement of particles. Financial research of the Faculty of Management, University of Tehran.
5) Peikarjou, K. (2009) Measuring the asset risk of companies and financial institutions using the value-at-risk method. Journal of Economic Research, 221-95.
6) Raei, R. (2006) Advanced investment management. Tehran: Samit Publications.
7) Sinaei, H. (2006) Decision making for stock portfolio selection, comparison of genetic and bee algorithms. Research paper on executive scientific research management.
8) Fazlzadeh, A. (2003) Investigating the ability of single-index Sharpe models and data coverage analysis in choosing efficient portfolios in Tehran Stock Exchange. Stock Exchange Quarterly.
9) Feizi, Zh. (2002) Investigating Monte Carlo methods for approximating efficiency at risk and conditional value at risk. The third financial and applied mathematics conference. Semnan: Semnan University.
10) Farahabadi, M. (2022 Using artificial intelligence network and Bayesian network model to predict liquidity risk in the banking industry. Stock Exchange Quarterly, (59) 15, 100-81.
11) Mehrjordi, Z. (2011) Hybrid intelligent algorithm based on mean-variance skewness fuzzy model for portfolio selection. International Journal of Industrial Engineering and Production Management.
12) Ajit Kumar Pasayat. (2023). Prediction based mean-value-at-risk portfolio optimization using machine learning regression algorithms for multi-national stock markets.
13) Bishop, T. (2000). variational relevance vector machines. proceedings of the sixteenth conference on uncertainty in artificial intelligence. Morgan kaufmann publishers .
14) Bustos. (2018). Multiobjective Genetic Programming, Redusing Bloat by using SPEA2. In Congress on Evolutionary Computation.
15) Chakraborty et al. (2018). A hybrid stock selection model using Genetic Algorithms and Support Vector Regression. Department of Computer Science and Information Engineerin.
15) Coyne. (2018). An Effective Decision Basic Genetic Algorithm Approach to Multiobjective Portfolio Optimization Problem. Applied Mathematical sciences.
17) Fischer and krauss. (2018). Robust Portfolio Optimization. john wiley.
18) Hu et al. (2018). the elements of statistical learning. soringer new york.
19) Hung et al. (2018). the elements of statistical learning, data mining. inference and prediction.
20) Hakimeh Morteza. (2023). An improved learning automata based multi-objective whale optimization approach for multi-objective portfolio optimization in financial markets.
21) Kia. (2018). A Double-Stage Genetic Optimization Algorithm for Portfolio Selection. ICONIP 2006.
22) Liu. (2018). data mining techniques. wiley.
23) Malagrino et al. (2018). Machine learning in sentiment reconstruction of the simulated stock market. statistical mechanics and its applications.
24) Mikhail Goykhman. (2019). A Novel Automatic satire and irony detection using ensembled feature selection and data mining. Elsevier.
25)Vu Minh Ngo. (2023). Does reinforcement learning outperform deep learning and traditional portfolio optimization models in frontier and developed financial markets?
26) Ren. (2018). Predicting the stock price of frontier markets using machine learning and modified Black–Scholes Option pricing model. statistical mechanics and its applications.
27) Reaz Chowdhury. (2001). A hybrid Genetic Quantitative Method for Risk-Return Optimization for Credit Portfolio Institute AIFB. Faculty of Economics.
28) Sotirios P. (2006). Portfolio Performance Evaluation in a mean-variance-skewness framework. European Journal of Operational Research.
29) Thomas Fischer. (2017). Deep adaptive group-based input normalization for financial trading. pattern recogniting letters.
30) Wang. (2018). Mean-Variance-skewness Model for Portfolio Selection With Transaction Costs. Information Journal of ystems Science.
31) wuyu, w. (2020). Portfolio formation with preselection using deep learning from long-term financial data. expert system with applications.
32) Zhang. (2018). Comparative analysis of expected shortfall and value at risk . institute for monetary and economic studies .
33) Zhou. (2018). support vector classification with input data uncertainty. advances in neural information processing systems.
34) Yaping Cai) 2019).A comprehensive cluster and classification mining procedure for daily stock market return forecasting. neurokomputing
35) Zolfaghari, M., & Gholami, S. (2020). A hybrid approach of adaptive wavelet transform, long short-term memory and ARIMA-GARCH family models for the stock index prediction. Expert Systems with Applications, 182, 115149
|_