Research and Analysis of Tesla Stock Price Prediction Based on ARIMA and LSTM Models

Faculty of Information Science and Technology, National University of Malaysia, Bangi, 43600, Malaysia

^* Corresponding author’s email: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

This study examines two time series forecasting methods: the Autoregressive Moving Average (ARIMA) model and the Long Short-Term Memory (LSTM) network. The main objective is to evaluate their accuracy and interpretability in predicting the Tesla stock market, while also reflecting their applicability to different market dynamics. This study uses daily closing price data for Tesla from 2010 to 2025 collected from Slick charts LLC. After normalization and differencing preprocessing, the data is divided into training and test sets. Experimental results show that the LSTM model outperforms the ARIMA model in both roots mean square error (RMSE) and mean absolute percentage error (MAPE). However, ARIMA remains more effective in predicting stable linear trends. In conclusion, this study finds that deep learning methods can improve the predictive performance of the Tesla stock market. Future research will explore a hybrid ARIMA-LSTM framework, combining the interpretability of the model with the nonlinear learning capabilities of neural networks to improve the stability of predictions.