Price Prediction and Feature Importance Analysis of German Electric Vehicles Based on Boosted Decision Tree Model
DOI:
https://doi.org/10.54691/bcpbm.v38i.4021Keywords:
Electric Vehicles; Price Prediction; Feature Importance; Boosted decision tree; Renewable Energy.Abstract
As people realize the importance of non-renewable energy sources, more and more industries are looking for alternative sources of energy that can be used. The automotive industry is a prime example. Electric vehicles are having an increasingly dominant role in the automotive market. This paper investigates the price of electric vehicles (EVs) and the importance of consumers' characteristics for electric vehicles in the European region. In addition, feature importance is discussed with disordered data in selecting the best regression model. Finally, the Boosted decision tree model predicts future electric vehicle prices in the German region. The results show that consumers place more importance on the size of the car, the battery's durability, and acceleration. Although the penetration of electric vehicles in Germany is still tiny, government incentives for electric vehicle companies and consumers will increase the market penetration of electric vehicles. So we believe that in the future, people will be able to buy good-performing electric cars at more affordable prices.
Downloads
References
Arababadi, A., Leyer, S., Hansen, J., & Arababadi, R. (2021). Characterizing the theory of spreading EVs in Luxembourg. Sustainability, 13(16), 9068.
Luchs, S. , & Kasperk, G. . (2017). The influence of government incentives on electric vehicle adoption: cross-national comparison. MAGKS Papers on Economics.
Ana Lucia Murillo. 2021. The Push for EVs Could Affect How Much Your Next Car Costs. https://money.com/electric-car-vs-gas-car-costs-biden/.
CC Jel. Does purchase price affect preferences for electric vehicle characteristics and environmental benefits?.
Gocompare. 2021. The future of electric vehicles: our predictions. https://www.gocompare.com/motoring/electric-cars/electric-car-adoption-prediction/.
Ev-database Organization. 2022. Compare EVs - EV Database (Version: 4.3). https://ev-database.org/.
Rodriguez, P. , Bautista, M. A. , Gonzalez, J. , & Escalera, S. . (2018). Beyond one-hot encoding: lower dimensional target embedding. Image and Vision Computing, 75(JUL.), 21-31.
Liz Najman. 2021. Measuring Efficiency In Electric Cars. https://blog.wa.aaa.com/green-technology/measuring-efficiency-in-electric-cars/.
Z. Zhang, K. T. McDonnell, E. Zadok and K. Mueller, "Visual Correlation Analysis of Numerical and Categorical Data on the Correlation Map," in IEEE Transactions on Visualization and Computer Graphics, vol. 21, no. 2, pp. 289-303, 1 Feb. 2015, doi: 10.1109/TVCG.2014.2350494.
A André, Laura, T. , Oliver, S. , & Thomas, L. . Permutation importance: a corrected feature importance measure. Bioinformatics(10), 1340.
Nabavi, & Jafari. (2013). Providing a customer churn prediction model using random forest technique. -, -(-), 202-207.






