Tools
Tools
Uluturk, Selin (2024) Regression Analysis for Predicting Prices of Used Cars: A Study Utilizing Data from Car Trading Website. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (1MB) | Preview |
Abstract
The aim of this study is to predict used car prices using various regression models. Car data collected from a used car trading website in Türkiye in June 2024 was used. The study focused on the fifteen best-selling car brands in Türkiye in 2023. The dataset was divided into four groups based on price, and each group was analyzed separately. The regression models used included Multiple Linear Regression (MLR), Support Vector Regression (SVR), Random Forest, and XGBoost. The analysis results showed that the XGBoost model had the highest R-squared values, while the Random Forest model had the lowest Mean Absolute Percentage Error (MAPE) values. Additionally, it was determined that gear type, specific car brands, and models played an important role in price prediction. This study aims to provide more accurate price predictions in the used car market, enabling both individual consumers and businesses to make more informed decisions.
Actions (login required)
 |
View Item |