A Feature-Engineered ARIMA-SARIMA Hybrid Model for Stock Price Prediction

Jialu Luo

2023

Abstract

Stock price prediction has long been a challenging yet vital task for investors and financial analysts. This research presents an innovative approach to enhance the accuracy of stock price predictions through the integration of the Feature-Engineered Auto-regressive Integrated Moving Average (ARIMA) and the Seasonal Auto-regressive Integrated Moving Average (SARIMA) hybrid model. The experiment and analysis reveal the superior predictive performance of the ARIMA-SARIMA hybrid model compared to standalone ARIMA or SARIMA models. By judiciously integrating seasonal and non-seasonal factors, the hybrid model mitigates the limitations of individual models in capturing the complex dynamics of stock price movements. This study opens new avenues for advancing stock price prediction models, offering investors and financial practitioners a valuable tool for making informed decisions in an increasingly complex and dynamic financial landscape. The fusion of traditional time-series analysis with feature engineering underscores the potential for more accurate and reliable stock price forecasts, with implications extending beyond financial markets to broader domains of time-series forecasting and prediction. Nevertheless, the fluctuations in stock prices are influenced by multiple factors, many of which lie beyond the predictive capability of existing models. Thus, while the hybrid model exhibits promising results, the author recognizes that further research is warranted to incorporate a broader spectrum of influential factors.

Paper Citation

in Harvard Style

Luo J. (2023). A Feature-Engineered ARIMA-SARIMA Hybrid Model for Stock Price Prediction. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 47-53. DOI: 10.5220/0012814600003885

in Bibtex Style

@conference{daml23,
author={Jialu Luo},
title={A Feature-Engineered ARIMA-SARIMA Hybrid Model for Stock Price Prediction},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={47-53},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012814600003885},
isbn={978-989-758-705-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - A Feature-Engineered ARIMA-SARIMA Hybrid Model for Stock Price Prediction
SN - 978-989-758-705-4
AU - Luo J.
PY - 2023
SP - 47
EP - 53
DO - 10.5220/0012814600003885
PB - SciTePress