A modular Python framework for my Bachelor Thesis at the University of Zurich, “Evaluating Trading Strategies in Cryptocurrency Futures Markets: A Comparative Analysis of Bitcoin and Ethereum.”

It provides a reproducible engine to download data, simulate trading strategies, and analyze performance on Binance perpetual futures (BTCUSDT, ETHUSDT).

.
├── Backtester.py           # Core engine: handles portfolio, trades, PnL computation
├── Bot_Class.py            # Trading bot wrapper combining data, strategy, and execution
├── Config_File.py          # Configuration file (API keys, parameters, paths)
├── TradingStrats.py        # Strategy implementations (EMA cross, RSI, ATR breakout, etc.)
├── trade_models.py         # Trade, Position, and Order object definitions
├── trade_utils.py          # Helper functions for trade handling and performance metrics
├── data_utils.py           # Data loading, cleaning, and resampling (Binance/CSV/parquet)
├── bulk_data_downloader.py # Binance historical data downloader
├── Helper.py               # General-purpose utilities (math, date/time, etc.)
├── visualization.py        # Plotting equity curves, drawdowns, and metrics
├── run_grid.py             # Batch/parameter sweep runner for strategy combinations
├── debug_backtester.py     # Debugging/testing harness
├── test_Trading_Strats.py  # Unit tests for strategy logic
├── logger.py               # Logging setup for console and file outputs
├── __init__.py             # Marks package
└── .gitignore              # Ignore venvs, data, and temporary logs

1. Clone the repository:

   git clone https://github.com/aki2273/TradingProgram.git

2. Create and activate a virtual environment:

   python -m venv .venv
   source .venv/bin/activate      # (Windows: .venv\Scripts\activate)

3. Install dependencies:

   pip install -r requirements.txt

Edit Config_File.py before running anything:

# Config_File.py
BINANCE_API_KEY = "your_api_key_here"
BINANCE_SECRET_KEY = "your_secret_key_here"

# General parameters
DATA_PATH = "data/"
RESULTS_PATH = "results/"
FEE_RATE = 0.0004
SLIPPAGE = 0.0001
LEVERAGE = 3

Fetch OHLCV and funding data from Binance Futures:

python bulk_data_downloader.py --pair BTCUSDT --interval 1h --start 2019-01-01 --end 2025-06-30

Data will be saved in data/ by default.

python Backtester.py --pair BTCUSDT --strategy ema_cross --start 2020-01-01 --end 2025-06-30

Example output:

Running EMA Cross strategy on BTCUSDT...
Final Equity: 1.284x
CAGR: 18.7%, Sharpe: 1.42, Max Drawdown: -22.5%
Results saved to: results/BTCUSDT_ema_cross_2025-06-30/

Explore multiple parameter combinations:

python run_grid.py --strategy rsi_band \
  --grid "rsi_len=[7,14,21]; rsi_low=[20,30]; rsi_high=[70,80]"

This will generate results for each combination and store metrics in a summary CSV.

Generate plots for equity curves, drawdowns, and rolling Sharpe ratios:

python visualization.py --input results/BTCUSDT_ema_cross_2025-06-30/

Each strategy implements:

def generate_signals(df, params):
    # returns long/short/flat signals as +1/-1/0

Run internal test scripts to verify logic:

python test_Trading_Strats.py
python debug_backtester.py

Each backtest produces a timestamped results folder under results/:

results/
└── BTCUSDT_ema_cross_2025-06-30_12-00-00/
    ├── trades.csv
    ├── daily_equity.csv
    ├── metrics.json
    ├── config_snapshot.json
    └── equity_plot.png

To add a new strategy:

1. Create a function in TradingStrats.py
2. Register it in the strategy dictionary inside Bot_Class.py
3. Pass your strategy name to Backtester.py

# TradingStrats.py
def bollinger_band(df, params):
    ...

All runtime logs (info, warnings, errors) are handled by logger.py. They’re stored under logs/{date}.log for debugging or audit purposes.

This project is for academic research only. It does not execute live trades and should not be used for financial decision-making. All Binance trademarks and data belong to their respective owners.

Akash Islam (2025) Evaluating Trading Strategies in Cryptocurrency Futures Markets: A Comparative Analysis of Bitcoin and Ethereum. Bachelor Thesis, University of Zurich (UZH), Department of Banking & Finance.

Akash Islam University of Zurich — Banking & Finance 📧 akash.islam@uzh.ch 💼 Research interests: Quantitative Finance, Algorithmic Trading, Machine Learning in Finance