A Machine Learning Approach for Study of Emission Standards on Used Car Prices in India
Kartikeya Mandhar
*
Wipro Technologies, India.
*Author to whom correspondence should be addressed.
Abstract
This study investigates the influence of emission standards on second-hand car prices in India using advanced machine learning techniques. Utilizing a comprehensive dataset from CarDekho, we performed two distinct analyses to explore this relationship. Initially, we excluded emission standards, employing various regression algorithms, with Random Forest and XGBoost achieving accuracies close to 94%. Upon introducing emission standards into the models, Random Forest's accuracy slightly improved to 94.25%, while XGBoost's accuracy decreased to 88.08%, highlighting different algorithmic responses to regulatory variables. These findings emphasize the critical role of emission standards in predictive modeling, offering valuable insights for policymakers and stakeholders in the automotive industry to make informed decisions that align with environmental objectives and market realities.
Keywords: Predictive modeling, emission standards, automotive economics, regression analysis, algorithm performance, machine learning, random forest, XGBoost, environmental regulations, market valuations