Skip to content

README.md

Latest commit

 

History

History
69 lines (47 loc) · 1.69 KB

README.md

File metadata and controls

69 lines (47 loc) · 1.69 KB
Raw

📊 Optiver Research — Trading at the Close

AI-powered stock closing price prediction using LightGBM on real Optiver Kaggle dataset.

Python LightGBM FastAPI React

🔗 Live Demo: optiver-trading.vercel.app

📸 Screenshot

Dashboard

🚀 Features

  • 🧠 LightGBM model trained on 500K real NASDAQ order book data
  • 📊 12 engineered features — WAP, Spread, Imbalance, Price Pressure
  • ⚡ FastAPI backend with real-time prediction
  • 📈 Feature Importance chart — interactive React dashboard
  • 🎯 Validation MAE: 5.1552

🛠️ Tech Stack

Layer Technology
ML Model LightGBM
Backend FastAPI
Frontend React + TypeScript
Charts Recharts
Data Optiver Kaggle Competition
Deployment Vercel + Render

⚙️ Setup

# Backend
cd backend
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt
uvicorn main:app --reload

# Frontend
cd frontend
npm install
npm start

📖 Usage

  1. Open optiver-trading.vercel.app
  2. Enter order book features (bid_price, ask_price, imbalance, etc.)
  3. Click Predict
  4. View predicted closing price + Feature Importance chart

👩‍💻 Author

Binary Mind — GitHub