The Research Of New Energy Vehicles Sales Forecasting In China

With the rapid development of China’s economy,the living standard of the residents has been greatly improved,and the automobile become the most popular way of transportation in our live.But at the same time,the environmental pollution is becoming more and more serious.License plate need to be approved by drawing lots,and the vehicle usage is limited by license plate number.As a result,people has to consider new energy vehicles(NEV).In order to solve the environment pollution and the energy shortage,our government announced policy to support the development of NEV industry.The market share of NEV is 14.8%in the year of 2021,compared with 5.8%in the year of 2020,increased by 9 percentage points.The NEV industry has become more and more important for the development of national economy and people’s daily life.The NEV sales is an important factor to measure the development trend of NEV.Giving a fairly accurate forecast of the NEV sales is significantly important for both macroeconomics and microeconomics.According to the periodic fluctuation of the NEV sales in China,we analyze the developing process,and make the both monthly and yearly sales forecasting.It can better help us control the market of NEV.In terms of the monthly sale forecasting,we have collected the sale data from Jan 2015 to Dec 2020,84 months.We use SARIMA and BP neural network to forecast monthly sales respectively.It shows that both SARIMA and BP neural network have a good predictive accuracy.To further improve the prediction effect,SARIMA and BP neural networks are combined to establish a hybrid forecasting model.It shows that the predictive capability of the hybrid forecasting model is better that the single model.In terms of the yearly sale forecasting,the private car ownership,GDP,the total highway mileage,PCDI,the Number of charging pile and the power purchasing price indices are used for sales forecasting.The yearly sales forecasting model is established at last.The results show that the mean absolute percentage error(MAPE)is 9.00,and the prediction effect is better.According to the growth of economy,the forecasting sales can be provided as standard version,optimism version and gloomy version for the 14th Five-Year Plan Period.