Lists
Lists
Ouazzane, Karim, Tang, Kai Hung Po Yung and Ghanem, Mohamed Chahine (2024) Optimal portfolio and trading strategy using machine learning. In: Global IEEE Congress on Emerging Technologies (GCET-2024), 9-11 December 2024, Gran Canaria, Spain. (In Press)
This research presents machine learning models for forecasting the future returns of a portfolio from NASDAQ semiconductors assets by financial analysis, optimization, and technical analysis to form a trading strategy. The performance of the portfolio is evaluated by back-testing. Data were collected from 2011 to 2019 for the sector of semiconductor companies listed on Nasdaq. The project consists of 4 sub-tasks. The first sub-task is to use the annual financial ratios of each company under the sector of semiconductors from 2011 to 2018 to project the company returns in 2019 using machine learning algorithms. Then, the top 5 highest-return assets would be selected to form a portfolio. After the optimization of the portfolio by Monte Carlo simulation, the classifiers adopt the technical indicators of the portfolio assets from 2011 to 2018 to predict the trading signals (buy or sell) in 2019. The trading actions in 2019 are simulated by back-testing. The result shows that the optimal portfolio using the simulat
2024338734.pdf - Accepted Version
Available under License Creative Commons Attribution 4.0.
Download (920kB) | Preview
 |
View Item |