Optimizing Stock Price Prediction Using Advan...

Optimizing Stock Price Prediction Using Advanced Machine Learning Algorithms: A Focus on XGBoost and Comparative Models

Author Name : Mohammad Aasim Shaikh, Nikhil Dharmendra Singh, Harsh C Vachheta, Kunal Randhir Sharma, Niraj Rampal, Atharva Vikram Salvi

DOI: https://doi.org/10.56025/IJARESM.2024.1211242724

Accurate stock price prediction is essential for financial analysts and investors to make informed decisions. This research explores the effectiveness of several machine learning models, including LSTM, KNN, Neural Network Regression, LightGBM and XGBoost in forecasting stock prices using historical data and market indicators such as trade volume, open, high, low and close prices. The models are evaluated using performance metrics like Mean Squared Error (MSE) and R-squared. The results highlight that the XGBoost model outperforms others achieving an exceptional R-squared value of 0.999702 and an MSE of 5.125808 demonstrating its superior accuracy. The findings suggest that XGBoost's ensemble learning approach effectively captures complex relationships in stock price data. This makes it a valuable tool for analysts seeking precision in market forecasting. Additionally, automating hyperparameter tuning using AutoML techniques could optimize the models for various datasets.