مدیریت مالی هوشمند

مدیریت مالی هوشمند

مروری نظام‌مند بر کاربرد روش‌های داده‌کاوی و یادگیری ماشین در ارزش‌گذاری سهام، همراه با یک چارچوب ترکیبی پیشنهادی برای پایداری صندوق‌های سهامی بورس تهران

نوع مقاله : مالی-بانکداری

نویسندگان
1 گروه حسابداری، واحد ارومیه، دانشگاه آزاد اسلامی، ارومیه، ایران
2 گروه حسابداری، دانشگاه ارومیه، ارومیه، ایران
چکیده
زمینه و هدف: پیش‌بینی قیمت سهام و ارزش‌گذاری آن همواره یکی از چالش‌های اصلی سرمایه‌گذاران و پژوهشگران حوزه مالی بوده است. با پیشرفت فناوری‌های هوش مصنوعی و داده‌کاوی، رویکردهای نوینی برای مدل‌سازی و پیش‌بینی ارزش سهام شکل گرفته است. هدف این پژوهش، مرور نظام‌مند تحقیقات انجام‌شده در زمینه کاربرد روش‌های داده‌کاوی و یادگیری ماشین در ارزش‌گذاری سهام و، بر پایه شکاف‌های شناسایی‌شده، ارائه یک چارچوب ترکیبی برای پایداری صندوق‌های سهامی بورس اوراق بهادار تهران است.
روش: این مطالعه از روش مرور نظام‌مند (Systematic Review) با رویکرد PRISMA بهره برده است. پایگاه‌های داده داخلی (پایگاه‌های علمی–پژوهشی ایران، مگیران و SID) و خارجی (ScienceDirect، Springer، IEEE و Google Scholar) در بازه زمانی 2000 تا 2024 جست‌وجو شدند و پس از غربالگری، 68 مقاله (28 داخلی و 40 خارجی) برای تحلیل نهایی انتخاب شدند.
یافته‌ها: روش‌های داده‌کاوی به پنج دسته اصلی تقسیم می‌شوند: (1) درخت تصمیم و مدل‌های مبتنی بر قانون، (2) شبکه‌های عصبی و یادگیری عمیق، (3) ماشین بردار پشتیبان، (4) روش‌های بیزین و (5) روش‌های ترکیبی و فراابتکاری. مدل‌های درخت تصمیم C5، CART و CHAID و شبکه‌های عصبی بیشترین کاربرد را در پیش‌بینی ارزش سهام داشته‌اند و روند تحقیقات از مدل‌های آماری سنتی به سمت یادگیری ماشین و یادگیری عمیق در حال حرکت است. همچنین نماگرهای نقدی نسبت به نماگرهای ساختاری و انتظاری از دقت پیش‌بینی بالاتری برخوردارند. بر پایه شکاف‌های شناسایی‌شده، یک چارچوب ترکیبی دومرحله‌ای شامل پیش‌بینی با شبکه عصبی Bi-LSTM و تابع زیان Huber و بهینه‌سازی استوار مبتنی بر CVaR پیشنهاد و به‌طور مقدماتی ارزیابی شد.
نتیجه‌گیری: با وجود گسترش پژوهش‌ها، همچنان شکاف‌هایی در زمینه مدل‌سازی جامع نماگرهای بازار سرمایه، استفاده از داده‌های کلان اقتصادی، مقایسه نظام‌مند روش‌ها و پیوند پیش‌بینی هوشمند با مدیریت ریسک وجود دارد. پژوهش حاضر با ارائه چارچوب طبقه‌بندی جامع و یک چارچوب ترکیبی پیشنهادی، مسیر تحقیقات آتی را ترسیم می‌کند.
کلیدواژه‌ها

عنوان مقاله English

A Systematic Review of the Application of Data Mining and Machine Learning Methods in Stock Valuation, Along with a Proposed Hybrid Framework for the Sustainability of Equity Funds in the Tehran Stock Exchange

نویسندگان English

Hossein Ashrafi Soltan Ahmadi 1
Saeed Jabarzadeh Kangarlouei 1
Jamal Bahri Sales 1
Ali Ashtab 2
1 Department of Accounting, Urmia Branch, Islamic Azad University, Urmia, Iran
2 Department of Accounting, Urmia University, Urmia, Iran
چکیده English

Background and Objective: Predicting stock prices and valuing stocks have consistently been major challenges for investors and researchers in the field of finance. With advancements in artificial intelligence and data mining technologies, novel approaches to modeling and forecasting stock value have emerged. The aim of this research is to systematically review existing studies on the application of data mining and machine learning methods in stock valuation and, based on identified research gaps, to propose a hybrid framework for the sustainability of equity funds in the Tehran Stock Exchange. Method: This study employs a systematic review methodology following the PRISMA approach. Domestic databases (Iranian Scientific-Research Databases, Magiran, and SID) and international databases (ScienceDirect, Springer, IEEE, and Google Scholar) were searched for the period 2000–2024; following a screening process, 68 articles (28 domestic and 40 international) were selected for final analysis.
Findings: Data mining methods are categorized into five main groups: (1) decision trees and rule-based models, (2) neural networks and deep learning, (3) support vector machines, (4) Bayesian methods, and (5) hybrid and meta-heuristic methods. Decision tree models (C5, CART, and CHAID) and neural networks have been the most widely used for stock value prediction, and the research trend is shifting from traditional statistical models toward machine learning and deep learning. Furthermore, cash-based indicators demonstrate higher prediction accuracy compared to structural and expectation-based indicators. Based on identified research gaps, a two-stage hybrid framework—comprising prediction using a Bi-LSTM neural network with a Huber loss function and robust optimization based on CVaR—was proposed and preliminarily evaluated. Conclusion: Despite the expansion of research, gaps remain regarding the comprehensive modeling of capital market indicators, the utilization of macroeconomic data, the systematic comparison of methods, and the integration of intelligent prediction with risk management. By presenting a comprehensive classification framework and a proposed hybrid framework, this study outlines the path for future research.

کلیدواژه‌ها English

Data mining
Machine learning
Stock valuation
Capital market indicators
Decision tree
Neural network
portfolio management
Conditional Value-at-Risk (CVaR)
آذر، ع.، و رجب‌زاده، ع. (۱۳۷۹). ارزیابی روش‌های پیش‌بینی ترکیبی با رویکرد شبکه‌های عصبی [پایان‌نامه کارشناسی ارشد]. دانشگاه تربیت مدرس.
ابراهیم‌نژاد، ع.، زمانی، ش.، و بیریا، م. (۱۳۹۹). پیش‌بینی رفتار قیمت سهام توسط دفتر سفارشات محدود: شواهدی از بورس اوراق بهادار تهران. مدیریت دارایی و تأمین مالی، ۸(۴)، ۶۳–۸۴.
ابونوری، ا.، و عبداللهی، م. (۱۳۹۱). مدلسازی نوسانات بخش‌های مختلف بازار سهام ایران با استفاده از مدل گارچ چندمتغیره. تحقیقات مالی، ۱۴(۱)، ۱–۱۶. https://doi.org/10.22059/jfr.2012.36628
افشاری، ح. (۱۳۸۲). بررسی ساختاری قابلیت پیش‌بینی قیمت سهام در بورس اوراق بهادار تهران. فصلنامه بررسی‌های حسابداری و حسابرسی، ۱۰(۱۲۶)، ۳۲–۱۰۳.
افلاطونی، ع. (۱۳۹۲). تجزیه و تحلیل آماری با EViews در تحقیقات حسابداری و مدیریت مالی. انتشارات ترمه.
البرزی، م.، یعقوب‌نژاد، ا.، و مقصود، ح. (۱۳۸۷). کاربرد شبکه‌های عصبی مصنوعی در پیش‌بینی شاخص بازدهی نقدی و قیمت سهام. مطالعات تجربی حسابداری مالی، ۶(۲۲)، ۱۱۹–۱۳۷.
علیزاده، س.، و ملک‌محمدی، س. (۱۳۹۳). داده‌کاوی و کشف دانش گام‌به‌گام با نرم‌افزار Clementine. انتشارات دانشگاه صنعتی خواجه نصیرالدین طوسی.
عرب‌صالحی، م.، و کمالی دهکردی، [نام کوچک درج نشده]. (۱۴۰۰). بررسی مقایسه‌ای رابطه‌های تنزیلی ارزش‌گذاری سهام در شرکت‌های پذیرفته‌شده در بورس اوراق بهادار تهران. چشم‌انداز مدیریت مالی، ۱۱(۳۳)، ۱۱۳–۱۳۸.
آشتاب، ع. — این نام در فهرست ارسالی شما به‌عنوان نویسنده منبع درج نشده است؛ بنابراین منبعی به آن اضافه نمی‌شود.
حاتم‌راد، س.، آدرنگی، ب.، اصغرپور، ح.، و حقیقت، ج. (۱۴۰۲). بررسی تأثیرگذاری عوامل کلان اقتصادی بر شاخص قیمت سهام بازار بورس ایران با استفاده از مدل‌های میانگین‌گیری. پژوهش‌های اقتصادی ایران، ۲۸(۹۵)، ۱۹۳–۲۳۶. https://doi.org/10.22054/ijer.2023.71810.1164
جهانگیری، خ.، و حسینی ابراهیم‌آباد، س. ع. (۱۳۹۶). بررسی آثار سیاست پولی، نرخ ارز و طلا بر بازار سهام در ایران. فصلنامه تحقیقات مالی، ۱۹، ۳۸۹–۴۱۴.
تهرانی، ر.، پیروز، ب.، و گورانی، [نام کوچک درج نشده]. (۱۳۹۳). امکان‌سنجی استفاده از مدل ترکیبی سهام (CANSLIM) در بورس اوراق بهادار تهران. فصلنامه دانش مالی تحلیل اوراق بهادار، ۷(۲۱)، ۲۹–۴۲.
کنی، ا. (۱۳۸۵). تحلیل تکنیکی پیشرفته. شرکت کارگزاری راهبرد سرمایه‌گذاری.
لشگری، ز.، و همکاران. (۱۳۹۷). رابطه بین رشد تولید ناخالص داخلی و عوامل ریسک سرمایه. تحقیقات حسابداری و حسابرسی، (۳۸)، ۹۵–۱۱۰.
روحی ملکی، س.، و پاک‌مرام، ع. (۱۳۹۶). بررسی رابطه محافظه‌کاری حسابداری با واکنش بازار و ارزش شرکت‌ها. حسابداری و منافع اجتماعی، ۷(۴)، ۹۹–۱۲۶. https://doi.org/10.22051/ijar.2017.9221.1093
سارنج، ع.، و نوراحمدی، م. (۱۳۹۵). تخمین ارزش در معرض ریسک (VaR) و ریزش مورد انتظار (ES) با استفاده از رویکرد ارزش فرین شرطی در بورس اوراق بهادار تهران. تحقیقات مالی، (۴۳).
صنیعی آباده، م.، محمودی، س.، و طاهرپرور، م. (۱۳۹۳). داده‌کاوی کاربردی. انتشارات نیاز دانش.
عباسیان، ع.، مرادپور اولادی، م.، و عباسیون، و. (۱۳۸۷). اثر متغیرهای کلان اقتصادی بر شاخص کل بورس اوراق بهادار تهران. فصلنامه پژوهش‌های اقتصادی ایران، ۱۲(۳۶)، ۱۲۸–۱۵۲.
Hou, X et al. (2020). An Enriched Time-Series Forecasting Framework for Long-Short Portfolio Strategy. IEEE Access, 8, 31992-32002. [https://doi.org/10.1109/ACCESS.2020.2973037](https://doi.org/10.1109/ACCESS.2020.2973037)
Sezer, O. B., Gudelek, M. U., & Ozbayoglu, A. M. (2020). Financial time series forecasting with deep learning: A systematic literature review: 2005–2019. Applied Soft Computing, 90, 106181. https://doi.org/10.1016/j.asoc.2020.106181
Kumbure, M. M., Lohrmann, C., Luukka, P., & Porras, J. (2022). Machine learning techniques and data for stock market forecasting: A literature review. Expert Systems with Applications, 197, 116659. https://doi.org/10.1016/j.eswa.2022.116659
Cont, R. (2001). Empirical properties of asset returns: Stylized facts and statistical issues. Quantitative Finance, 1(2), 223–236. https://doi.org/10.1080/713665670
Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. The Review of Financial Studies, 33(5), 2223–2273. https://doi.org/10.1093/rfs/hhaa009
Fernandez, P. (2002). Valuation methods and shareholder value creation. Academic Press.
Fernandez, P. (2004). The value of tax shields is NOT equal to the present value of tax shields. Journal of Financial Economics, 73(1), 145–165. [https://doi.org/10.1016/j.jfineco.2002.10.001](https://doi.org/10.1016/j.jfineco.2002.10.001)
Bicer, B. (2006). Stock valuation for Investment and corporate decisions. ProQuest.
Appel, G. (2005). Technical analysis: Power tools for active investors. FT Prentice Hall.
Achelis, S. B. (2001). Technical analysis from A to Z (2nd ed.). McGraw-Hill.
Ahmed, I. (2018). Impact of dividend per share and earnings per share on stock prices: A case study from Pakistan (textile sector). International Journal of Social Sciences, Humanities and Education, 2(2), 1–10.
Banz, R. W. (1981). The relationship between return and market value of common stocks. Journal of Financial Economics, 9(1), 3–18. https://doi.org/10.1016/0304-405X(81)90018-0
Deng, L., & Yu, D. (2014). Deep learning: Methods and applications. Foundations and Trends in Signal Processing, 7(3–4), 197–387. https://doi.org/10.1561/2000000039
Markowitz, H. (1952). Portfolio selection. The Journal of Finance, 7(1), 77–91. https://doi.org/10.1111/j.1540-6261.1952.tb01525.x
Goldfarb, D., & Iyengar, G. (2003). Robust portfolio selection problems. Mathematics of Operations Research, 28(1), 1–38. https://doi.org/10.1287/moor.28.1.1.14260
Cornuejols, G., & Tütüncü, R. (2006). Optimization methods in finance. Cambridge University Press.
Rockafellar, R. T., & Uryasev, S. (2000). Optimization of conditional value-at-risk. Journal of Risk, 2(3), 21–41. https://doi.org/10.21314/JOR.2000.038
Quaranta, A. G., & Zaffaroni, A. (2008). Robust optimization of conditional value at risk and portfolio selection. Journal of Banking & Finance, 32(10), 2046–2056. https://doi.org/10.1016/j.jbankfin.2007.12.025
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
Fischer, T., & Krauss, C. (2018). Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research, 270(2), 654–669. https://doi.org/10.1016/j.ejor.2017.11.054
Siami-Namini, S., Tavakoli, N., & Namin, A. S. (2019). The performance of LSTM and BiLSTM in forecasting time series. In 2019 IEEE International Conference on Big Data (pp. 3285–3292). IEEE. https://doi.org/10.1109/BigData47090.2019.9005997
Zhang, Z., Zohren, S., & Roberts, S. (2020). Deep learning for portfolio optimization. The Journal of Financial Data Science, 2(4), 8–20. https://doi.org/10.3905/jfds.2020.1.042
Li, X., & Wu, P. (2022). Stock price prediction incorporating market style clustering. Cognitive Computation, 14(1), 149–166. https://doi.org/10.1007/s12559-021-09820-1
Deng, Y., Hope, O.-K., Wang, C., & Zhang, M. (2022). Capital market liberalization and auditors' accounting adjustments: Evidence from a quasi-experiment. Journal of Business Finance & Accounting, 49(1–2), 215–248. https://doi.org/10.1111/jbfa.12541
Elmghaamez, I. K., Attah-Boakye, R., Adams, K., & Agyemang, J. (2022). The diffusion of innovation theory and the effects of IFRS adoption by multinational corporations on capital market performance: A cross-country analysis. Thunderbird International Business Review, 64(1), 81–108. https://doi.org/10.1002/tie.22243
Sadorsky, P. (2022). Forecasting solar stock prices using tree-based machine learning classification: How important are silver prices? The North American Journal of Economics and Finance, 61, 101705. https://doi.org/10.1016/j.najef.2022.101705
Ang, A., Chen, J., & Xing, Y. (2006). Downside risk. The Review of Financial Studies, 19(4), 1191–1239. [https://doi.org/10.1093/rfs/hhj035](https://doi.org/10.1093/rfs/hhj035)
Soni, P., Tewari, Y., & Krishnan, D. (2022). Machine learning approaches in stock price prediction: A systematic review. Journal of Physics: Conference Series, 2161, 012065. https://doi.org/10.1088/1742-6596/2161/1/012065
Göçken, M., Özçalıcı, M., Boru, A., & Dosdoğru, A. T. (2016). Integrating metaheuristics and artificial neural networks for improved stock price prediction. Expert Systems with Applications, 44, 320–331. https://doi.org/10.1016/j.eswa.2015.09.029
Lim, B., & Zohren, S. (2021). Time-series forecasting with deep learning: A survey. Philosophical Transactions of the Royal Society A, 379(2194), 20200209. https://doi.org/10.1098/rsta.2020.0209
Huber, P. J. (1964). Robust estimation of a location parameter. The Annals of Mathematical Statistics, 35(1), 73–101. https://doi.org/10.1214/aoms/1177703732
Alipour, M. A., & Parsinejad, S. (2018). Designing an appropriate neural network to assist in more accurate stock portfolio predictions. Transactions on Machine Intelligence, 1(4), 228–242.
[https://doi.org/10.47176/TMI.2018.228](https://doi.org/10.47176/TMI.2018.228)
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71
Quinlan, J. R. (1993). C4.5: Programs for machine learning. Morgan Kaufmann.
Hu, H., Tang, L., Zhang, S., & Wang, H. (2018). Predicting the direction of stock markets using optimized neural networks with Google Trends. Neurocomputing, 285, 188–195. https://doi.org/10.1016/j.neucom.2018.01.038
Chang, C.-C., & Lin, C.-J. (2011). LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2(3), Article 27. https://doi.org/10.1145/1961189.1961199
Qiu, Y., et al. (2022). “Short-term stock trends prediction based on sentiment analysis and machine learning.” Soft Computing, 26, 2209–2224. [https://doi.org/10.1007/s00500-021-06602-7](https://doi.org/10.1007/s00500-021-06602-7)
Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. In Proceedings of the 3rd International Conference on Learning Representations (ICLR). https://doi.org/10.48550/arXiv.1412.6980