NYC Real Estate Price Predictor

Features

• Predicts price per square foot and total price for properties
• Covers all five NYC boroughs
• Accounts for multiple factors:
  • Property size (gross square feet)
  • Land size
  • Building age
  • Building type (Residential/Commercial/Other)
  • Location (Borough)
• User-friendly web interface built with Streamlit
• Model trained on NYC Department of Finance rolling sales data

Model Performance

• Average prediction error: ~15%
• Key predictive features:
  • Property size
  • Location (Borough)
  • Building age
  • Land-to-building ratio

Installation

1. Clone the repository:

   git clone https://github.com/geoffgin/Real_Estate_Pricing.git
   cd Real_Estate_Pricing

2. Install required packages:

   pip install -r requirements.txt

Usage

1. Train the model:

   python model_training.py

2. Run the web interface:

   streamlit run app.py

3. Open your browser and go to http://localhost:8501

Project Structure

nyc-real-estate-predictor/
├── assets/
│   ├── nyc-rolling-sales.csv
│   └── streamlit_interface.png
├── model_training.py
├── app.py
├── nyc_price_model.pkl
├── requirements.txt
└── README.md

Data

The model is trained on the NYC Department of Finance Rolling Sales Data, which includes:

• Property sales prices
• Building characteristics
• Location information
• Usage type

Model Details

• Algorithm: XGBoost Regressor
• Features engineered:
  • Log transformations of square footage
  • Building age categories
  • Commercial ratio
  • Land-to-building ratio
  • Units per square foot

GitHub Repository

geoffgin/Real_Estate_Pricing