FinancePy: End-of-Day Market Pipeline

A robust, offline-first financial analysis pipeline for daily market summaries, calculating advanced indicators (Hurst, Market Mode, Volatility) and generating Markdown reports.

Features

• Data Ingestion: Robust yfinance fetching with local caching (no re-download).
• Indicators: Bollinger Bands, EWMA Volatility, Hurst Exponent (DFA), Market Mode (PCA).
• Analytics: Regime classification (Trend/Mean Revert) and Risk-Off logic.
• Reporting: Daily markdown summary with watchlist and systemic risk analysis.

Structure

• src/: Core logic (data, indicators, signals, reporting).
• tests/: Unit and smoke tests.
• configs/: Configuration (tickers, parameters).
• reports/: Generated daily reports.
• data_cache/: Local Parquet store.

Installation

1. Clone repo.
2. Environment:

   # Create venv (optional but recommended)
   python -m venv venv
   source venv/bin/activate  # Linux/Mac
   venv\Scripts\activate     # Windows
   
   # Install dependencies
   pip install pandas numpy yfinance pyyaml scipy

Usage

Daily Run

Execute the pipeline for the current date (or specific date):

python run_daily.py --date 2025-12-30

Reports are saved in reports/.

Quality Assurance (Tests)

To validate the codebase (Unit tests + Smoke test):

# Option 1: Using Make (if available)
make qa

# Option 2: Using Python script (Cross-platform)
python scripts/qa.py

Dashboard

Launch the interactive visualizer (Streamlit):

make dashboard
# or
streamlit run app/dashboard.py

Troubleshooting

• Missing Data: Check logs/ or console output. The pipeline warns but continues.
• NaN in Indicators: Hurst requires sufficient history. Market Mode requires multiple assets.
• Network Issues: The pipeline uses cached data if available. First run requires internet.

Architecture

See docs/ARCHITECTURE.md for details on data flow and design contracts.