USING ARTIFICIAL INTELLIGENCE TO SOLVE THE OPTIMAL STOP TIME PROBLEM IN FINANCIAL INVESTMENT

Van Khanh Pham, Thanh Trung Nguyen

Main Article Content

Abstract

In this paper, we present an advanced tool of Artificial intelligence, reinforcement learning, to test in stock investing. Artificial intelligence basically includes machine learning, deep learning and reinforcement learning. Reinforcement learning uses mathematical theories such as dynamic programming, Markov decision processes to improve actions to become more optimal. Reinforcement learning has many different algorithms, in this article we use Zap Q-Learning algorithm to apply in investing 30 stocks of Vietnam stock market. Our results are quite modest: after discounting the bank interest, the profit is about 3%.

Article Details

References

[1] Chen, S., Devraj, A. M., Busic, A., Meyn, S., (2019), Zap Q-Learning for Optimal Stopping Time Problems, arXiv:1904.11538v3.
[2] Choi, D. and Van Roy, B., (2006), A generalized Kalman filter for fixed point approximation and efficient temporal-difference learning, Discrete Event Dynamic Systems: Theory and Applications, 16(2):207–239.
[3] Sutton, R. S. and Barto, A. G., (2018), Reinforcement Learning: An introduction, The MIT Press, Cambridge, Massachusetts.
[4] Tsitsiklis, J. N. and Van Roy, B., (1999), Optimal stopping of Markov processes: Hilbert space theory, approximation algorithms, and an application to pricing high-dimensional financial derivatives, IEEE Trans. Automat. Control, 44(10):1840–1851.
[5] Đặng Hùng Thắng, (2007), Giáo trình xác suất: Quá trình ngẫu nhiên và tính toán ngẫu nhiên, NXB Đại học quốc gia Hà Nội, Hà Nội.
[6] Nguyễn Duy Tiến, (2000), Các mô hình xác suất và ứng dụng: Phần I – Xích Markow và ứng dụng, NXB Đại học quốc gia Hà Nội, Hà Nội.