Predicted Used Car Price in the UK Using Linear Regression and Machine Learning Models

Dexin Huang

2025

Abstract

The global used car market has witnessed substantial growth, particularly in Europe, driven by extensive cross-border trading and improved vehicle quality. Accurate price prediction in this market is critical for reducing information asymmetry, enhancing transparency, and supporting informed consumer decisions. This study compares three regression models-Multiple Linear Regression (MLR), Random Forest (RF), and K-Nearest Neighbours (KNN)-to predict used car prices in the UK, utilizing a comprehensive dataset of 3,685 vehicles sourced from Kaggle. Extensive feature engineering was applied, including log transformation of vehicle prices, imputation of missing values, and encoding of categorical variables, to enhance model performance. Results indicate that the Random Forest model achieved the highest predictive accuracy, yielding a Coefficient of Determination (R²) value of 0.8714 and the lowest Mean Absolute Error (MAE) of £821.76. While MLR provided valuable interpretability, it suffered from multicollinearity issues, and the KNN model underperformed in high-dimensional settings. These findings highlight critical methodological insights and practical implications, contributing to more accurate, data-driven pricing strategies in the UK’s automotive resale market.

Paper Citation

in Harvard Style

Huang D. (2025). Predicted Used Car Price in the UK Using Linear Regression and Machine Learning Models. In Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy - Volume 1: IAMPA; ISBN 978-989-758-774-0, SciTePress, pages 534-540. DOI: 10.5220/0013829000004708

in Bibtex Style

@conference{iampa25,
author={Dexin Huang},
title={Predicted Used Car Price in the UK Using Linear Regression and Machine Learning Models},
booktitle={Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy - Volume 1: IAMPA},
year={2025},
pages={534-540},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013829000004708},
isbn={978-989-758-774-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy - Volume 1: IAMPA
TI - Predicted Used Car Price in the UK Using Linear Regression and Machine Learning Models
SN - 978-989-758-774-0
AU - Huang D.
PY - 2025
SP - 534
EP - 540
DO - 10.5220/0013829000004708
PB - SciTePress