Stock Market Forecasting Using Machine Learning Models Through Volatility-Driven Trading Strategies

Ivan Letteri

2024

Abstract

The purpose of our research was to explore volatility-based trading strategies in financial markets to leverage market dynamics for capital gain. We sought to introduce a strategy that integrated statistical analysis with machine learning to predict stock market trends. Our method involved using the k-means++ clustering algorithm to examine the mean volatility of the nine largest stocks in both the NYSE and Nasdaq markets. The clusters formed the basis for understanding relationships among stocks based on their volatility patterns. We further subjected the mid-volatility clustered dataset to the Granger Causality Test, which helped identify stocks with strong predictive connections. These stocks were crucial in formulating our trading strategy, serving as trend indicators for decisions on target stock trades. Our empirical approach included thorough backtesting and performance analysis. Our findings demonstrated the effectiveness of our method in exploiting profitable trading opportunities. This was achieved through predictive insights derived from volatility clusters and Granger causality relationships among stocks. In conclusion, our research contributed to the field of volatility-based trading strategies by offering a methodology that combined a statistical approach with machine learning. This enhanced the predictability of stock market trends.

Paper Citation

in Harvard Style

Letteri I. (2024). Stock Market Forecasting Using Machine Learning Models Through Volatility-Driven Trading Strategies. In Proceedings of the 6th International Conference on Finance, Economics, Management and IT Business - Volume 1: FEMIB; ISBN 978-989-758-695-8, SciTePress, pages 96-103. DOI: 10.5220/0012607200003717

in Bibtex Style

@conference{femib24,
author={Ivan Letteri},
title={Stock Market Forecasting Using Machine Learning Models Through Volatility-Driven Trading Strategies},
booktitle={Proceedings of the 6th International Conference on Finance, Economics, Management and IT Business - Volume 1: FEMIB},
year={2024},
pages={96-103},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012607200003717},
isbn={978-989-758-695-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 6th International Conference on Finance, Economics, Management and IT Business - Volume 1: FEMIB
TI - Stock Market Forecasting Using Machine Learning Models Through Volatility-Driven Trading Strategies
SN - 978-989-758-695-8
AU - Letteri I.
PY - 2024
SP - 96
EP - 103
DO - 10.5220/0012607200003717
PB - SciTePress