Predicting the U.S. Stock Market Index Using LSTM with Different Financial Factors

Tingxi Zhang

2024

Abstract

The complexity and dynamic nature of financial markets demand advanced tools for accurate forecasting. This is vital for investors, portfolio managers, and policymakers to make informed decisions regarding asset allocation and risk management. This study explores the potential of Long Short-Term Memory (LSTM) networks in predicting the S&P 500 index, augmented by a diverse set of financial factors including the Cboe Volatility Index (VIX), Effective Federal Funds Rate (EFFR), U.S. Dollar Index (USDX), and various U.S. Treasury rates. The research employs an approach involving data collection spanning from June 2010 to June 2023, preprocessing to ensure data suitability, and LSTM model development tailored to capture long-term dependencies. This article starts from two tasks, classification and regression, and focuses on predicting the S&P 500 index across varying time horizons. The study reveals that LSTM models augmented with relevant financial factors effectively predict short-term movements in the S&P 500 index, with low Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values for 1-day predictions. However, prediction errors increase significantly for longer horizons, particularly for variables highly sensitive to market volatility and interest rate changes. The findings contribute to developing more accurate forecasting tools for the financial industry, enhancing decision-making capabilities for various stakeholders.

Paper Citation

in Harvard Style

Zhang T. (2024). Predicting the U.S. Stock Market Index Using LSTM with Different Financial Factors. In Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI; ISBN 978-989-758-726-9, SciTePress, pages 251-256. DOI: 10.5220/0013214300004568

in Bibtex Style

@conference{ecai24,
author={Tingxi Zhang},
title={Predicting the U.S. Stock Market Index Using LSTM with Different Financial Factors},
booktitle={Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI},
year={2024},
pages={251-256},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013214300004568},
isbn={978-989-758-726-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on E-commerce and Artificial Intelligence - Volume 1: ECAI
TI - Predicting the U.S. Stock Market Index Using LSTM with Different Financial Factors
SN - 978-989-758-726-9
AU - Zhang T.
PY - 2024
SP - 251
EP - 256
DO - 10.5220/0013214300004568
PB - SciTePress