Tesla stock prediction and analysis based on LSTM model

Ziyu Geng
DOI: https://doi.org/10.54254/2753-8818/26/20241020
2023-12-20
Abstract:Now that new energy vehicles are developing well, Teslas stock forecast has research value. This report focuses on predicting and analysing Tesla stock price returns using Long Short-Term Memory (LSTM) models. Deep learning models like LSTM can handle large amounts of data and make predictions about future stock dynamics. In this research, historical stock prices of Tesla Inc. are utilized as input data. The LSTM model is used to train and test the data, and subsequently provides results on its accuracy. For comparison, both Linear Regression and Random Forest models have also been used. The results indicate that the LSTM model has better performance than the other models in predicting short-term stock price movements. The result is evaluated by MSE, MAE and RMSE. However, Stock prices are extremely susceptible to economic, market, and political factors, so the predictions of the LSTM model cannot play an important role in actual investment.
What problem does this paper attempt to address?