📈 Financial Time Series Analytics Platform

A full-stack financial analytics dashboard featuring real-time stock market data analysis, ARIMA forecasting models, and interactive visualizations—containerized with Docker and deployable to AWS EC2 with custom domain support.

🌟 Overview

This platform provides comprehensive time series analysis and forecasting for financial markets. It combines statistical modeling (ARIMA), interactive visualizations, and a modern web interface into a containerized application ready for cloud deployment.

Key Achievement: Solved complex Docker volume mounting and composition challenges for seamless data persistence across environments—from development to production on AWS EC2.

✨ Features

📊 Analytics & Modeling

• Real-time Stock Data: Live market data fetching via Yahoo Finance API
• ARIMA Forecasting: Time series prediction with configurable parameters (p, d, q)
• Statistical Analysis:
  • Moving averages and volatility metrics
  • Returns analysis and distribution modeling
  • Autocorrelation and partial autocorrelation functions
  • Seasonal decomposition (trend, seasonal, residual)
• Normality Testing: Shapiro-Wilk and visual Q-Q plots

📈 Interactive Visualizations

• Multi-panel Dashboard: Candlestick charts, volume analysis, price movements
• Technical Indicators: Rolling means, Bollinger bands, volatility measures
• Forecast Visualization: Model predictions with confidence intervals
• Diagnostic Plots: Residual analysis, ACF/PACF plots, decomposition charts

🎨 User Interface

• Modern Streamlit Dashboard: Responsive layout with intuitive controls
• Ticker Selection: Quick-select popular stocks or enter custom symbols
• Date Range Filtering: Analyze specific time periods
• Customizable Visualizations: Adjustable parameters and color schemes
• Real-time Updates: Dynamic data refresh on selection changes

🏗️ Architecture & Deployment

• Dockerized Application: Containerized setup for easy deployment
• AWS EC2 Deployment: Production-ready cloud infrastructure
• Custom Domain: Cloudflare DNS integration for professional access
• Data Persistence: Volume mounting for environment consistency
• CI/CD Ready: Makefile automation for builds and deployments

🚀 Quick Start

Prerequisites

# System Requirements
- Python 3.8+
- Docker & Docker Compose
- Git

Local Development

1. Clone the repository

git clone https://github.com/yourusername/architecture-fullstack.git
cd architecture-fullstack

2. Set up Python environment

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

3. Run the dashboard locally

streamlit run main.py

The dashboard will open at http://localhost:8501

Docker Deployment

1. Build and run with Docker Compose

docker-compose up --build

2. Access the application

• Dashboard: http://localhost:8501

Using the Makefile

# Build new image
make build_new

# Build and publish to Docker Hub
make drhub

# Build and publish to GitHub Container Registry
make git

📁 Project Structure

architecture-fullstack/
├── main.py                 # Main Streamlit application
├── datamover.py           # Data fetching and preprocessing utilities
├── requirements.txt       # Python dependencies
├── Dockerfile            # Container image definition
├── docker-compose.yaml   # Container orchestration
├── Makefile             # Build and deployment automation
├── dev_folder/          # Development notebooks and data
│   ├── dev_notebook.ipynb
│   └── timeseries_analysis.ipynb
└── .streamlit/          # Streamlit configuration
    └── config.toml

🛠️ Technology Stack

Core Technologies

• Python 3.8+: Primary programming language
• Streamlit: Interactive web application framework
• Pandas & NumPy: Data manipulation and numerical computing
• yfinance: Yahoo Finance API wrapper for real-time market data

Analytics & Modeling

• statsmodels: ARIMA time series forecasting
• SciPy: Statistical analysis and testing
• scikit-learn: Additional ML utilities

Visualization

• Plotly: Interactive charts and graphs
• Matplotlib & Seaborn: Statistical visualizations
• mplfinance: Financial-specific chart types (candlesticks, volume)

DevOps & Infrastructure

• Docker: Containerization platform
• Docker Compose: Container orchestration
• AWS EC2: Cloud compute instance
• Cloudflare: DNS and CDN services

📊 Usage Examples

Analyzing a Stock

1. Select a ticker from the sidebar (AAPL, GOOG, MSFT, TSLA) or enter a custom symbol
2. Choose your date range using the date pickers
3. Explore the visualizations:
   • Historical price data and volume
   • Statistical metrics and distributions
   • ARIMA forecast predictions
   • Decomposition analysis

ARIMA Model Configuration

The dashboard allows you to adjust ARIMA parameters:

• p (AR order): Number of lag observations
• d (Differencing): Degree of differencing
• q (MA order): Size of moving average window

Forecast Horizon

Set the number of days to forecast into the future (1-90 days).

🏛️ Architecture

Application Architecture

┌─────────────────┐
│   User Browser  │
└────────┬────────┘
         │
    ┌────▼────────┐
    │  Streamlit  │ (Port 8501)
    │  Dashboard  │
    └────┬────────┘
         │
    ┌────▼────────┐
    │  yfinance   │
    │     API     │
    └─────────────┘

Data Flow

1. Data Ingestion: Real-time stock data fetched via Yahoo Finance API
2. Processing: Data cleaning, transformation, and feature engineering
3. Analysis: Statistical analysis and ARIMA model training
4. Visualization: Interactive charts rendered in Streamlit
5. Deployment: Containerized application ready for cloud deployment

🌐 Deployment to AWS

EC2 Setup

1. Launch EC2 Instance

   • AMI: Ubuntu 20.04 LTS
   • Instance Type: t2.micro or larger
   • Security Group: Open port 8501

2. Install Docker

sudo apt-get update
sudo apt-get install docker.io docker-compose
sudo usermod -aG docker ubuntu

3. Deploy Application

git clone <your-repo>
cd architecture-fullstack
docker-compose up -d

4. Configure Domain (Optional)
   • Point your domain to the EC2 instance IP in Cloudflare DNS

🧪 Development

Running Jupyter Notebooks

The dev_folder/ contains Jupyter notebooks for experimentation:

cd dev_folder
jupyter notebook

Adding New Features

1. Create a new branch
2. Implement your feature in main.py or create new modules
3. Test locally with streamlit run main.py
4. Update tests and documentation
5. Submit a pull request

🐛 Troubleshooting

Common Issues

Port already in use

# Kill process on port 8501
lsof -ti:8501 | xargs kill -9

Docker volume issues

# Remove all containers and volumes
docker-compose down -v
docker system prune -a

yfinance data not loading

• Check your internet connection
• Verify ticker symbol is valid
• Yahoo Finance may temporarily rate-limit requests

📝 Configuration

Streamlit Configuration

Edit .streamlit/config.toml to customize:

• Theme colors
• Server settings
• Browser behavior

Docker Configuration

Modify docker-compose.yaml to adjust:

• Port mappings
• Volume mounts
• Resource limits

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request. For major changes:

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Yahoo Finance for providing free financial data API
• Streamlit for the amazing web framework
• statsmodels for time series analysis tools

📧 Contact

Dominik Huffield
Twitter - @cogentdom
My Work - portfolio
Business - ohmic data

Project Link: architecture-fullstack

---

⭐ If you found this project helpful, please consider giving it a star!