For the XGBoost model, I selected key features (year, mileage, and Population) and converted the data into the required format for training. After fitting the model, I evaluated its performance, and it gave the best results among all models with an RMSE of $4,463, MAE of $3,208, and an R-squared of 0.86, meaning it explained 86% of the variation in car prices. This shows that XGBoost was highly effective at capturing the patterns in the data and making accurate predictions.