Philadelphia Housing Price Prediction
Improving Property Tax Assessments
Isabelle Li, Luciano Lu, Sihan Yu
Introduction
- Why this matter
- What we did
- Goal
Data Sources
| Source | Description | Key Variables / Features |
|---|---|---|
| OPA Property Sales | 2023–2024 residential transactions | sale_price, total_livable_area, number_of_bedrooms, number_of_bathrooms, house_age, garage_spaces |
| Census ACS 2023 | Socioeconomic indicators by tract | median_income, ba_rate, unemployment_rate, total_pop |
| OpenDataPhilly – Transit | Public transit locations | Count of stops within 400 m buffer (transit_count) |
| OpenDataPhilly – Recreation / Parks | Recreation locations | Count within 1200 m buffer (recreation_count), categorical dummy (recreation_dummy) |
| OpenDataPhilly – Crime | Crime incidents | Count of incidents within 400 m (crime_count) |
| Hospitals | Hospital locations | Distance to nearest hospital (nearest_hospital_m) |
| Center City boundary | Geojson polygon | Binary dummy (center_city_dummy) |
Data Overview
Where Are Expensive Homes?
What Drives Prices?
Top Predictors
Model Performance
Hardest to Predict
Recommendations
Ensure Fair and Transparent Property Valuation
Build Community Resilience through Economic Revitalization
Promote Spatial Equity in Access to Opportunity
Limitations & Next Steps
Limited Performance on High-End Properties
Spatial but Not Temporal
Data Incompleteness and Systematic Bias
Questions?
Thank You!
Contact: isabli@upenn.edu · GitHub: [https://github.com/Isabelliiii]
Contact: yusihan@upenn.edu · GitHub: [https://github.com/sihan-yu429]
Contact: lu25@upenn.edu · GitHub: [https://github.com/lluluciano0505]