-Forecasting Chinese Electric Vehicle Volume Using Keyword Search Data Based on SARIMA-LSTM Model

China has become the world's largest market for electric vehicles. This article proposes a hybrid forecasting method that combines SARIMA(Seasonal Autoregressive Integrated Moving Average) model with LSTM(Long Short-Term Memory) neural network to address the issue of electric vehicle volume. The SARIMA-LSTM model is combined with network keyword search index to improve its accuracy in predicting electric vehicle volume. The article first uses SARIMA model to make a preliminary linear prediction of the trend of electric vehicle volume, and then uses LSTM neural network to further optimize the SARIMA prediction results, improving the accuracy of the prediction. Through the comparison of SARIMA models with and without accelerated network search index, it is found that adding network search index can improve the prediction accuracy of SARIMA. Finally, the SARIMA, LSTM, and SARIMA-LSTM models are compared, and it is found that the SARIMA-LSTM model has the highest accuracy. Simulation experiments show that this method has the best prediction effect.