Advancements in Communication and Systems
Unlocking Market Trends: LSTM-based Stock Price Forecasting for Intelligent Investments
Authors: Jaya Bharathi M and Ilavarasan S
Publishing Date: 13-09-2024
ISBN: 978-81-955020-7-3
Abstract
This paper introduces an advanced Long Short-Term Memory (LSTM) neural network model designed to predict future stock prices, effectively addressing the challenges of the non-linear and dynamic nature of financial markets. By integrating Explainable AI (XAI) techniques, specifically LIME (Local Interpretable Model-agnostic Explanations), with LSTM networks, we enhance both predictive accuracy and model interpretability. Our optimized LSTM model achieves a notable 92% accuracy in forecasting stock price trends, outperforming baseline autoregressive integrated moving average (ARIMA) models, and demonstrates robust performance with an impressive RMSE of 0.052, showcasing its capability to capture complex market dynamics. The incorporation of LIME provides transparent insights into the factors influencing individual predictions, enhancing trust and validation of the model’s decisions. Additionally, we introduce a comprehensive evaluation framework that combines traditional performance metrics with XAI-driven insights, offering a holistic assessment of model robustness and interpretability. Utilizing a dataset of historical daily stock prices for multiple companies and employing a sliding window approach to create input-output pairs, we conduct extensive experimentation with various network architectures, optimization algorithms, and input data representations to identify an optimal LSTM configuration for this task. This research not only advances the field of stock price prediction but also addresses the critical need for explainable AI in financial forecasting, laying the groundwork for developing more transparent and reliable data-driven algorithms for stock valuation and algorithmic trading with potential applications across diverse financial markets and asset classes
Keywords
CSV, RMSE, MAPE, RNN, Stock Prediction, LSTM, Time Series, Financial Forecasting, Deep Learning, Tensorflow, and Keras serve as essential anchors for researchers seeking relevant literature
Cite as
Jaya Bharathi M and Ilavarasan S, "Unlocking Market Trends: LSTM-based Stock Price Forecasting for Intelligent Investments", In: Ashish Kumar Tripathi and Vivek Shrivastava (eds), Advancements in Communication and Systems, SCRS, India, 2024, pp. 627-633. https://doi.org/10.56155/978-81-955020-7-3-55