Predicting Stock Fluctuations: Comparative Analysis of Advanced Machine Learning Models Using Tesla Stock
Download as PDF
DOI: 10.23977/ICEMESS2023.113
Author(s)
Mingjie Zhu, Yuan Cheng, Lin Huang, Ningjia Duan
Corresponding Author
Mingjie Zhu
ABSTRACT
This essay explores the use of advanced machine learning models to predict stock fluctuations, focusing on Tesla's stock performance. It emphasizes the advantages of these models in handling complex datasets and non-linear relationships. The experiment evaluates three models: Random Forest Regressor, LSTM, and XGBoost, using historical stock prices and sentiment analysis of Twitter data. The results show that XGBoost performs the best, followed by LSTM, while Random Forest Regressor exhibits lower accuracy. XGBoost and LSTM align more frequently with real stock trends, while Random Forest matches only a small percentage. Overall, this essay provides valuable insights into the application of advanced machine learning models for predicting Tesla's stock trends, benefiting traders, analysts, and investors in the stock market.
KEYWORDS
Deep Learning, LSTM, XG Boost, Random Forest, Sentimental Analysis