Algorithmic Trading Bot

This project is an implementation of a laboratory work for the course "Software Bots and Agents for Stock Markets." It demonstrates the development and testing of an algorithmic trading strategy.

Run My Project in Google Colab

You can run my project in Google Colab by clicking the link below:

Project Structure

The project is implemented in Google Colab and includes the following stages:

1. Data Access/Cleaning: Fetching and cleaning historical market data for SOL-USD from Yahoo Finance.
2. Pre-trade Analysis: Analyzing asset properties and defining indicators (SMA, ATR, RSI, Volatility) for the trading strategy.
3. Trading Signal & Trade Execution: Developing the AdvancedSmaStrategy trading strategy based on moving averages, RSI, with dynamic position sizing and SL/TP. Backtesting and parameter optimization using the backtesting.py library and the sambo method.
4. Post-trade Analysis: Analyzing strategy performance using metrics from backtesting.py and a detailed report generated by quantstats.

Key Results

• Strategy Total Return: +176% (vs. +31% for Buy & Hold SOL-USD).
• Compound Annual Growth Rate (CAGR): +39.34% (vs. +9.15% for Buy & Hold).
• Max Drawdown: -19.65% (vs. -91.88% for Buy & Hold).
• Sharpe Ratio: 1.44 (vs. 0.6 for Buy & Hold).

The strategy demonstrated a significant outperformance of the benchmark with substantially lower risks.

How to Run

1. Open the .ipynb file in Google Colab.
2. Install the necessary libraries (usually done at the beginning of the notebook via !pip install ...).
3. Run the cells sequentially.

Future Directions

• Implement adaptive SL/TP based on ATR.
• Integrate market sentiment analysis and fundamental factors.
• Conduct extensive testing, including out-of-sample periods and other assets.