Private car ownership forecasting considering the factor of new energy vehicles

Abstract

The rapid development of new energy vehicles and ride-hailing has injected many new uncertainties into the change of private car ownership. Reasonable and accurate prediction of the development trend of car ownership is helpful for the state to formulate relevant policies and measures. Based on private car ownership in China from 2000 to 2021, statistical data, considering the traditional and non-traditional factors selection such as per capita GDP, residents' consumption level, a total of 12 factors, using Principal Component Analysis (PCA) dimensionality reduction processing, establishing the principal component and the Logistic regression model between private car ownership. Finally, the prediction results of PCA-Logistic regression model were compared with the results obtained by using the cubic exponential smoothing method. The research results showed that the introduction of new factors such as ride-hailing and new energy vehicles, Using PCA-Logistic regression model can improve the prediction accuracy to a certain extent.