ارائه مدلی برای پیش‌بینی ارزش‌گذاری معاملات بلوکی با رویکرد شبکه عصبی مصنوعی LSTM

محورهای موضوعی : بازارها و نهادهای مالی

عادله بحرینی ¹ , مریم اکبریان فرد ^{2 *} , مهدی خوشنود ³

1 - گروه مهندسی مالی، واحد رشت، دانشگاه آزاد اسلامی، رشت، ایران.
2 - گروه حسابداری، واحد صومعه سرا، دانشگاه آزاد اسلامی، صومعه سرا، ایران.
3 - گروه حسابداری، واحد رودسر و املش، دانشگاه آزاد اسلامی، رودسر، ایران.

تاریخ دریافت : 1402/04/25 تاریخ پذیرش : 1402/08/15 تاریخ انتشار : 1402/09/07

کلید واژه: نوسانات قیمت, ارزش‌گذاری معاملات بلوکی, شبکة عصبی یادگیری عمیق, مدلLSTM,

چکیده مقاله :

هدف: در غالب پژوهش‌های پیشین در ایران و سایر کشورهای دیگر توانمندی سیستم‌های هوشمند در پیش‌بینی متغیرهای اقتصادی و مالی به‌ویژه قیمت سهام تأیید شده است، اما در ارزش‌گذاری معاملات بلوکی برای اولین‌بار محاسبه می‌گردد. هدف پژوهش حاضر بررسی نتایج رویدادها و اطلاعات از گزارش‌‌های مالی شرکت‌های پذیرفته شده در بورس اوراق بهادار تهران در قالب 15 شاخص مالی و یافتن میزان تأثیرگذاری این شاخص‌ها بر ارزش‌گذاری معاملات بلوکی با استفاده از آزمون Rmse بر روی داده‌های Test موردمطالعه قرار گرفته است.روش‌شناسی پژوهش: بدین منظور از اطلاعات مالی 64 شرکت ازمجموعه شرکت‌های پذیرفته شده درسازمان بورس اوراق بهادارتهران برای دوره زمانی 1390 تا1400 استفاده شده است. فرضیه‌ی تحقیق با بهره‌گیری از شبکه عصبی یادگیری عمیق مدل LSTM آزمون شده‌است.یافته‌ها: شبکة عصبی LSTM به جهت توانمندی بالا در آموزش داده‌ها و وزن‌های مناسب به این داده‌ها و خلق مسیری که با سرعت و دقت نتایج قابل‌قبولی جهت پیش‌بینی ارزش‌گذاری معاملات بلوکی دارد.اصالت / ارزش‌افزوده علمی: در مدل ارائه شده با اندازه‌گیری ارزش‌گذاری معاملات بلوکی، قیمت این معاملات، اثرات اطلاعات و نقدینگی معاملات با اندازه‌بزرگ را واپایش خواهیم نمود.

چکیده انگلیسی:

Objective: The capability of intelligent systems in predicting economic and financial variables, particularly stock prices, has been confirmed in previous research in Iran and other countries. However, the valuation of block transactions is calculated for the first time in this study. The aim is to investigate the outcomes and information from the financial reports of listed companies on the Tehran Stock Exchange using 15 financial indices and determine the impact of these indices on the valuation of block transactions by employing the RMSE test on the Test dataset.Research Methodology: For this purpose, financial information from 64 companies within the accepted companies of the Tehran Stock Exchange for the period from 1390 to 1400 has been utilized. The research hypothesis is tested using the Long Short-Term Memory (LSTM) deep learning neural network model.Findings: The LSTM neural network, due to its high capability in training data, appropriate weights for these data, and creating a path that efficiently and accurately produces acceptable results for predicting the valuation of block transactions.Originality/Value: In the proposed model, by measuring the valuation of block transactions, we will scrutinize the prices of these transactions and the effects of information and liquidity in large-sized transactions.

منابع و مأخذ:

Ahmadpour, A., & Nasiri, M. (2016). Surveying Price impact of block trades in the Iran stock market. Financial Research Journal, 18(1), 23-38. [In Persian]

Albuquerque, R. & Schroth, E. (2010). Quantifying private benefits of control from a structural model of block trades. Journal of Financial Economics, 96(1), 33-55.

Alikhani, A., & Yar Afshana, S. (2022). comparison of different types of profits and their effect on excess stock return, quarterly journal of judgment and decision making in accounting and auditing; types of profit and their effect on stock returns2(5), 39-58. [In Persian]

Alizadeh, H., Zanjirdar, M., & Haji, G. A. (2023). Ability of Machine Learning Algorithms and Artificial Neural Networks in Predicting Accounting Profit Information Content Before Announcing. Advances in Finance and Investment, 4(2), 1-30. [In Persian]

Ameri, M., & Belgurian., M. (2016), Thesis on Damrani's investigation of large and block transactions in the Tehran Stock Exchange market, p. 5.

Amini Mehr, A., & Hekmat, H. (2023). the power of convolutional neural network in predicting financial helplessness, Financial Management Strategy Quarterly, Year 11, Number 41. 77-96. [In Persian]

 Asadi, M., Mirbarg, M., & Chirani, E. (2022). presenting a neural network model for predicting the profits of companies listed on the Tehran Stock Exchange and comparing its accuracy with HDZ and ARIMA models, Scientific Journal of Management Accounting, 3(54), 163-180. [In Persian]

Basu, N., Paeglis, I., & Toffanin, M. (2017). Reading between the blocks. Journal of Corporate Finance, 8(45), 294-317.

Benaissa, N. (2023). Effect of Block Trading on the Moroccan Stock Exchange,VOL 5 NO, 33-52.

Breuker, D., Delfmann, P.,Matzner, M., & Becker, J. (2016). Comprehensible Predictive Models for Business Processes. MIS Quarterly, 40(4),1009-1034.

Comerton-Forde, C., & Putninš, T. J. (2016). Dark trading and price discovery. Journal of Financial Economics 118, 70-92.

David, A., & Jonathan, P. (2019). Market Trends Block Trades, Lexis Practice Advisor.2-5.

De, S. and J. Jindra, (2012). Why newly listed firms become acquisition targets. Journal of Banking & Finance, 36(9). 2616-2631.

Dehghan Nistanki, M., Izdi, H., & Alidousti, F. (2016). Investigating the impact of corporate governance indicators on the rate of return on assets of Tehran Stock Exchange banks, International Conference on New Researches in Management, Economics and Accounting, Period: 4.

Dong, L., Uchida, M., & Hou, X. (2014). Block trade targets in China. Journal of Corporate Finance, (25), 188-201.

Etamadi, H., Dehghani, T., Azar, A., Anwari, R., & Asghari A. (2013). Designing a model for pricing control blocks of stocks. Scientific Research Quarterly Journal of Financial Knowledge of Securities Analysis, 6(18), 71-84. [In Persian]

Fama, E ., & French, K. (1993). Common Risk Factors In The Returns On Stocks and Bonds. Journal of Financial Economics, Vol. 33. 3-56.

Ghasemi, D. S., Asgari, H., & Pakmaram, A. (2018). Experimental test of the information content of Ayers and Olson's adjusted cash flow model in companies listed on the Tehran Stock Exchange, National Conference on Management, Accounting and Economics with an emphasis on marketing Regional and global 4, number of pages 15. [In Persian]

Gul Arzi, Gh. H., & Badi Dast, I. (2021). Investors perception of market returns and its effect on the formation of collective behavior with an approach based on collective adherence to beta, Research Paper on Development and Capital, Volume 7, Number 1 - Serial Number 12. 87-100. [In Persian]

Harris, M., & Raviv, A. (1988). Corporate control contests and capital structure. Journal of financial Economics,. 20: 55-86. [In Persian]

Hatami, N. (2018), combining neural networks for stock forecasting. Journal of Economic Sciences, 10. 61-80.

Hosani, M., & Moradi, S. (2019). The mediator role of asymmetric information in the imperfect competition market on the relationship between profit forecast bias and specific risk resulting from the capital asset pricing model, Journal of Financial and Investment Developments,1(1), 81-100. [In Persian]

Ibadi, A. M., Jaghari, M. A ., & Davoudi Far, N. (2021). Stock price forecasting in the financial market using combined GA-SVM algorithms, Journal of Advances in Finance and Investmen,2(5), 1-22. [In Persian]

Jensen, M.C. (1986). Agency costs of free cash flow, corporate finance, and takeovers. The American economic review,76(2), 323-329.  

Lakshmi, B., & Gomini, P. (2003). Stock Market Volatility. A Survey of North America, Europe and Asia, National University of Singapore, Department of Economics, 21.

Mehrani, S., Moradi, M., Iskandar, H., & Hashemi Mir Mohammad, J. (2014). Institutional Ownership and Financial Flexibility, Scientific Research, 7(28), 43-56.

Mehraban, M. R., Tehrani, R., & Jamshidi, H. (2017). Analysis of the role of block transactions in creating abnormal returns and the effect on unsystematic fluctuations in the Tehran Stock Exchange, Scientific Quarterly of Asset Management and Financing, 1 (28), 2. [In Persian]

Nielsen., & Michael, A. (2015). NeuralNetworks and Deep learnin. Determination Press.

Osmani, F., Cheshmi, A., Salehnia, N., & Ahmadi, Shadmehr, M. (2023). The response of stock returns of various Iranian industries to inflation and interest rates with the Panel-ARDL approach, Planning and Budget Research Quarterly, 28(1), 75 -53. [In Persian]

Qingbin, M., Xuan, S., Chunlin, L., Qun, W., & Hongchao, Z. (2020). The Impact Of Block Trades On Stock Price Synchronicity From China, Sciencedirect, 68(2), 239-253.

Refk, S. (2018). Walid Mensi, Shawkat Hammoudeh, Jamal Bouoiyour Energy Economics bay shears a hedge, a safe haven or a diversifier for price movements.

Salehnejad, H., & Ghayor, V. (2019) the effect of the rate of return on assets and the rate of return on equity and the financial leverage of the shares of companies admitted to the Tehran Stock Exchange, Publication Researcher, 7(18), 17- 27. [In Persian]

Sediqi, A. H., & Sajdinejad, A. (2018). presenting an approach based on deep learning to detect fraud in financial payment systems, Information Management Scientific Quarterly, 5(1). 458-458. [In Persian]

Syedkhani, R., Mohammadi, M. A., & Amini P. (2021). Investigating the ability of operational cash flows in assessing the performance of companies with an emphasis on the quality of disclosure during periods of financial crisis, Journal  Researches in Financial Accounting and Auditing, 13(49), 147-176. [In Persian]

Shamsi, R., Gholami., & Jamkarani, R. (2022). Disclosure of information in the capital market, challenges and harms, Journal of Financial and Investment Developments, Spring 1-28. [In Persian]

Tehrani, R., Flowerjani., & Ruholah, G. (2007). Examining the ratio of book value to market value as a risk substitute variable using the leverage approach, Journal of Accounting and Auditing, 15(52), 37-54. [In Persian]

Vazan, M. (2020). the book Principles, Concepts and Approaches of Deep Learning, Miyad Andisheh Publications, 29-33. [In Persian]

Liu, Y. (2020). The Impacts of Dark Trading and Block Trading on Firm Valuation and Default Risk, Opal.latrobe.edu.au.

Sun, Y., & Ibikunle, G. (2015). Informed Trading and the Price Impact of Block Trades. SSRN Electronic Journal.Nikoomaram, H., Pourzamani, Z., & Dehghan, A. Spillover Effect On The On Contest Markets For Capital Market. Journal of Investment Knowledge, 3(11), 179-200.

Zhangkai, H., & Xingzhong, X. (2002). The Pricing Of Block Shares: Trading Off Liquidity Against Control,National Science Fund, 5(27) 1-19.

Zhang, G. P. (2003). Time Series Foricasting Using A Hybrid Arima And Neural Network Model, Neurocompuling, 3(50), 159-175.

Zare, M. H., & Nilchi, M. (2018). Comparative evaluation of Markowitz approach with a hybrid method to form an optimal portfolio using DNN deep learning and gravity search algorithm, Shahid Beheshti University, Financial Management Perspective Journal, 7(28), 165-188. [In Persian]

Zulfiqari, M., Sahabi, B., & Bakhtiaran M. (2019). Designing a model to predict the returns of the total index of the stock market (with an emphasis on the hybrid models of deep learning network and family models) GARCH, Quarterly, Journal of Financial Engineering and Securities Management, 9 (42), 138-169. [In Persian]