Iguchi Yukihiro    Department   Undergraduate School  , School of Science and Technology    Position   Professor
Date	2023/01/07
Presentation Theme	A Strategy for Stock Tradingusing Machine Learning
Promoters	IEICE
Conference Type	Workshop/Symposium
Presentation Type	Speech (General)
Contribution Type	Collaborative
Country	Japan
Holding period	2023/01/07～2023/01/08
Publisher and common publisher	Rintaro YODA, Tsutomu SASAO, and Yukihiro IGUCHI
Details	Inthisstudy, we propose a stock trading method that outperforms conventional methods in termsofshort-term trading returns by using stock price data and machine learning. In the stock price forecasting, we predict whether the stock price will go up, stay, or down. A low-passfilter is applied to the stock price waveform, and a trading strategy suitable for the wave form is applied based on machine learning predictions. The target stocks are TOPIX, which is composed of the major components of stocks listed on the Tokyo Stock Exchange, and the TOPIX Sector Indices. In the 10-year trading simulation for TOPIX, the sum of the rates of return is 27.22% higher than that of the conventional method.