Prediction of the Development of China's New Energy Industry

Abstract

In the past decade, China's new energy vehicle industry has developed rapidly due to the influence of national policies. In recent years, China has started planning to cease policy support for the new energy vehicle industry. Therefore, predicting the future development of China's new energy industry is of great significance for policy-making in China. This paper selects 10 factors influencing the development of new energy electric vehicles in the past ten years and 4 indicators reflecting the development of new energy electric vehicles. We combine the Entropy Weight Method with TOPSIS to score the development status of new energy electric vehicles based on 4 indicators. Then, we use Principal Component Analysis to reduce the dimensionality of the ten influencing factors to three principal components. To quantify these effects, Lasso regression is performed using 3 principal components to prevent overfitting. Therefore, this article obtained a model for predicting the development status of China's new energy vehicle industry, with a goodness of fit of over 0.98, indicating a good fit of the model. In order to predict the development score for the next decade, this paper first predict the data of ten factors for the next decade. The factors that remain unchanged are not treated. Factors that change the trend significantly are predicted using GM (1,1) and the forecast is good. For stationary time series, this paper collects data from nearly 20 years and trains on Long Short-Term Memory Networks. R2 = 0.891 and RMSE < 0.1, Therefore, the model can be used to predict data for the next 10 years. After obtaining various data, this paper obtains the development score of new energy electric vehicles in the next 10 years according to the regression equation. Finally, this paper concludes that the development and progress of China's new energy industry will slow down in the next decade, with a score increase of about 20%.