Predicting Used Vehicle Prices
Sarp Ata Kanca, Yagmur Beren Sengezken
ECON 465 Stage 3
Predicting Used Vehicle Prices
Authors
Sarp Ata Kanca
Yagmur Beren Sengezken
Economic Question
Research Question
To what extent do vehicle age, mileage, condition, and engine size predict the market price of a used vehicle?
Why Does It Matter?
- Used vehicle markets are economically important.
- Consumers need accurate price expectations.
- Sellers need efficient pricing strategies.
- Vehicle depreciation affects household wealth.
Dataset
Source
Kaggle: Craigslist Cars & Trucks Dataset
Sample
- Original observations: 426,880
- Cleaned sample: 170,439 observations
Variables
Price: Vehicle Listing Price , Odometer:Miles Driven , Cylinders:Engine Size
Year: Manufacturing Year , Condition:Vehicle Condition
Price Distribution
![]()
Observation
- Strong right-skewed distribution.
- Extreme high-value listings present.
- Log transformation improves model performance.
Model A
Baseline Regression
[ log(price)=_0+_1(year)+_2(odometer)]
Economic Intuition
- Newer vehicles should be worth more.
- Higher mileage should reduce value.
Model B
Extended Regression
[ log(price)=_0+_1(year)+_2(odometer)+_3(condition)+_4(cylinders)]
Why Add These Variables?
- Vehicle quality matters.
- Engine size affects consumer demand.
- More realistic representation of market valuation.
Model Comparison
| RMSE |
0.8344 |
0.7392 |
| R?? |
0.1383 |
0.3237 |
Result
Model B clearly performs better.
- Lower prediction error
- Higher explanatory power
Cross-Validation Results
5-Fold Cross Validation
Interpretation
- Model performance is stable across samples.
- Results generalize well to unseen observations.
Key Findings
Vehicle Age
- Positive coefficient
- Newer vehicles command higher prices
Odometer
- Negative coefficient
- More mileage lowers market value
Condition
- “Like New” vehicles receive significant premiums
- Salvage vehicles receive substantial discounts
Cylinders
- Engine size contributes to valuation differences
Economic Interpretation
Depreciation
Vehicle prices decrease as usage increases.
Consumer Valuation
Buyers pay premiums for:
- Newer vehicles
- Better condition
- Larger engines
Market Implications
Observable quality characteristics strongly influence market outcomes.
Limitations
Limitation 1
Important variables were omitted:
- Manufacturer
- Model
- Fuel type
- Location
Limitation 2
Data is self-reported by sellers.
Listing prices may differ from final transaction prices.
Future Research
Potential Improvement
Include:
- Brand effects
- Fuel efficiency
- Geographic market differences
New Economic Question
Do vehicle brands create a measurable price premium after controlling for age, mileage, condition, and engine size?
Conclusion
Main Takeaways
Vehicle age and mileage are significant predictors.
Condition substantially improves prediction accuracy.
Model B outperforms Model A.
Results are consistent with economic theory of depreciation.
Thank You
Questions?