Car Price Regression Analysis

A concise, reproducible project to model and explain used-car prices. We explore features (age, mileage, brand, engine, fuel, transmission, etc.), run baseline and regularized regression models, and compare performance (RMSE/MAE/R²). The repository includes a short report summarizing findings.

---

Goals

• EDA: understand distributions, outliers, and correlations.
• Feature engineering: encode categorical variables, create age/mileage transforms, handle missing values.
• Modeling: compare linear and regularized baselines (e.g., Ridge/Lasso) and optionally tree-based regressors.
• Evaluation: RMSE/MAE/R² on hold-out or cross-validation; simple error plots and feature importance.

---

Repository structure

. ├─ Data/ # raw or prepared datasets (place files here; see Data/README if added) / ├─ Report/ # project report (PDF/Docx) and any exported figures/tables / └─ README.md /

Data policy. If the dataset is from a public source (e.g., Kaggle), do not commit large/raw files; instead, link instructions to download in Data/README.md. If small, anonymized samples are allowed, you can include a 100–1,000 row sample for quick testing.

---

Quick start

This project can be run in Python or R depending on your scripts/notebooks. Keep one path and delete the other once you finalize.

Run in ** R **

1. Install packages:

  pkgs <- c("tidyverse","janitor","skimr","GGally","caret","glmnet","rpart","ranger","ggplot2")
  to_install <- setdiff(pkgs, rownames(installed.packages()))
  if(length(to_install)) install.packages(to_install)

2. Put data in Data/.

3. Knit the analysis Rmd or run the R script(s).

Reproducibility checklist

Random seeds fixed during CV/splits.

Data versions documented (source URL, download date).

Environment pinned (Python requirements.txt or R renv.lock).

Outputs: keep a few key plots/tables inside Report/ so visitors see results without running code.

Results (short preview)

Top predictors typically include age and mileage, with brand/segment fixed effects improving fit.

Regularization (Ridge/Lasso) often improves generalization vs. plain OLS; tree-based models can capture non-linearities (check SHAP or permutation importance for interpretability).

Replace this section with your actual metrics table (RMSE/MAE/R²) and 1–2 plots exported to Report/.

How to cite

If you use this repo:

Jinmin Li. Sicheng Huang. Zixiang Xiao. Yifan Wang (2024). Car Price Regression Analysis. Repository: https://github.com/Lexie000/Car_price_regression_analysis