Residual Value Calculation of New Energy Used Vehicles Based on XGBoost+Feature Engineering

At present, the new energy used car industry urgently needs a standard method that can accurately calculate the residual value of new energy used cars. This paper first determines the product calculation dimension based on the data from Guazi Website; Secondly, we need to perform data preprocessing such as searching and processing for outliers and missing values in the data source; Once again, we need to perform feature engineering processing such as deleting, merging, and reorganizing data dimensions; Finally, the XGBOOST model was used to calculate the residual value of new energy used cars. The results showed that when the sample size of popular car models was high enough, the model's prediction results could reach over 80%. The model and method proposed in this paper can meet the needs of industry development in calculating popular car models.