metadata
title: Dynamic Allocation System
emoji: 📈
colorFrom: blue
colorTo: green
sdk: docker
app_file: app.py
pinned: false
Dynamic Portfolio Allocation System
A machine learning-based portfolio allocation system that combines supervised learning and reinforcement learning to generate dynamic stock allocation recommendations.
Technologies Used
- Python
- Streamlit
- Gradient Boosting Regression
- PPO Reinforcement Learning
- Plotly
- Scikit-learn
- Stable-Baselines3
Features
- Dynamic portfolio allocation
- Risk profile selection
- 5-day and 10-day investment horizons
- Cash reserve optimization
- Portfolio visualization
- PPO vs Equal Weight comparison
- Portfolio insights
- Download portfolio as CSV
Project Workflow
- Feature Engineering
- Gradient Boosting Return Prediction
- PPO Reinforcement Learning Allocation
- Portfolio Recommendation
- Interactive Dashboard
Developed as a Final Year Project.