Time Series Forecasting and Correlation Analysis of the New Energy Electric Vehicle Industry Using ARIMA Algorithm and LSTM Neural Networks

This study embarks on an analytical journey to forecast the NEV industry, leveraging advanced time series models including the Autoregressive Integrated Moving Average (ARIMA) algorithm and Long Short-Term Memory (LSTM) recurrent neural networks. In a comparative analysis, employing t-tests to discern the correlation between datas. To ensure the robustness of the findings, distribution tests are applied to the datasets, validating the accuracy of the results. With the Kruskal-Wallis H test utilized to evaluate the effects of resistance on the future trajectory. The research in this paper provides new ideas for subsequent research on time series forecasting and helps to analyze the fluctuations and correlations of time series more accurately.