A Comparative Study of the Effectiveness of Machine Learning for Predicting Stock Prices in the New Energy Vehicle Industry

This paper investigates the effectiveness of machine learning models in predicting stock prices in the New Energy Vehicle (NEV) industry - a key part of the current transformation of the global transport system and electricity supply. This study applies five models, ARIMA, LSTM, GRU, CNN and TCN using the sample 20 NEV companies, to predict stock prices. And further use the risk metrics such as MSE, RMSE, R2, MAE and MAPE to measure the prediction performance. The results shows that the CNN model stands out for its superior accuracy in predicting stock movements, highlighting its potential as a valuable tool for investors and financial analysts in the fast-growing NEV market. The findings highlight the importance of utilizing advanced neural network algorithms to improve investment strategies and risk assessment in the face of market volatility and economic fluctuations.