Algo Trading System

A comprehensive algorithmic trading system in Python that generates buy/sell signals based on market regimes and technical/volatility strategies, sending notifications directly to a Telegram bot.

Features

• Data Fetching: Robust downloading and local caching of stock data using yfinance.
• Regime Detection: Combines Hidden Markov Models (HMM) and GARCH volatility forecasting to classify the market as TRENDING, RANGING, or VOLATILE.
• Strategies:
  • Trend Following: EMA Crossover, SuperTrend, MACD, Price Breakout.
  • Mean Reversion: RSI Extremes, Bollinger Bands, Z-Score Deviation.
  • Volatility: GARCH Vol Forecast, ATR Expansion, Volatility Mean Reversion.
• Risk Management: Position sizing based on Kelly Criterion and fractional risk, along with ATR/trailing stop losses.
• Backtesting Engine: Vectorized backtesting for fast optimization and Walk-Forward Analysis (WFA) to ensure out-of-sample robustness.
• Signal Delivery: Asynchronous Telegram bot integration to receive daily summaries and actionable trade alerts.

Installation

1. Clone the repository (if applicable) and navigate to the directory:

   cd algo-trading

2. Install dependencies:

   pip install -r requirements.txt

   (Note: TA-Lib is required by pandas-ta for some indicators. You may need to install the TA-Lib C library on your system first. On Mac: brew install ta-lib)

3. Configure Environment: Copy the example environment file and fill in your details:

   cp .env.example .env

   Edit .env to add your Telegram bot token and chat ID.

Setup Telegram Bot

1. Open Telegram and search for @BotFather.
2. Send /newbot and follow the prompts to create a new bot.
3. Copy the bot token (e.g., 123456:ABC-DEF1234ghIkl-zyx57W2v1u123ew11) and add it to your .env file as TELEGRAM_BOT_TOKEN.
4. To get your Chat ID:
   • Send any message to your new bot.
   • Visit: https://api.telegram.org/bot<YOUR_TOKEN>/getUpdates
   • Find "chat":{"id": YOUR_CHAT_ID} in the JSON response.
   • Add it to your .env file as TELEGRAM_CHAT_ID.

Usage

The system is controlled via the main.py CLI.

1. Fetch Data Fetch historical data for the US market (or specify tickers):

python main.py fetch
python main.py fetch --market ID
python main.py fetch --tickers AAPL MSFT TSLA

2. Detect Regime Analyze the current market regime for a specific stock:

python main.py regime --ticker AAPL

3. Generate Signals Run the full pipeline (data -> regime -> strategy -> risk) and generate signals based on your risk profile:

python main.py signal --profile aggressive
python main.py signal --profile conservative --notify

(Use --notify to send the generated signals to your configured Telegram bot).

Exploration

A set of Jupyter notebooks is provided in the exploration/ directory to help you understand the components, visualize data, build strategies, and run detailed backtests.

Project Structure

• config/: System settings, risk profiles, and universes.
• data/: Data fetching and caching layer.
• indicators/: Technical and fundamental indicators using pandas-ta.
• regime/: Statistical tests, HMM, and GARCH models.
• strategies/: Implementation of all trading strategies.
• risk/: Position sizing, stop loss, and portfolio tracking.
• backtesting/: Backtest engine, WFA, and metrics.
• signals/: Signal orchestrator and Telegram bot delivery.
• exploration/: Jupyter notebooks for research.