ptahlmud 0.0.9

Trading strategies crafting tools

from ptahlmud.core import Fluctuations

ptahlmud

Python library for crafting and backtesting trading strategies. Ptahlmud helps you design, test, and evaluate algorithmic trading strategies using historical market data.

Features

This project is still under construction and features may be added or reworked.

• Signal-based trading: Define entry and exit points with customizable signals
• Portfolio simulation: Track your strategy's performance over time
• Risk management: Configure position sizing, take-profit, and stop-loss parameters
• Backtesting engine: Test strategies against historical market data

Installation

pip install ptahlmud

Quick Start

Here's a simple example of defining trading signals and running a backtest:

from datetime import datetime
from pathlib import Path

from ptahlmud.backtesting.models import Signal, Side, Action
from ptahlmud.backtesting.backtest import TradingParameters, process_signals
from ptahlmud.datastack import FluctuationsService, BinanceClient, FluctuationsSpecs

client = BinanceClient(binance_secret="", binance_key="")
service = FluctuationsService(client, savedir=Path("/data/fluctuations_data"))

specifications = FluctuationsSpecs(
    coin="BTC", currency="USDT",
    from_date=datetime(2023, 1, 1),
    to_date=datetime(2023, 2, 1),
)

# Fetch and load the historical market data
service.fetch(specifications)
fluctuations = service.request(specifications)

# Define trading signals
signals = [
    Signal(date=datetime(2023, 1, 1), side=Side.LONG, action=Action.ENTER),
    Signal(date=datetime(2023, 1, 15), side=Side.LONG, action=Action.EXIT),
    Signal(date=datetime(2023, 2, 1), side=Side.SHORT, action=Action.ENTER),
    Signal(date=datetime(2023, 2, 15), side=Side.SHORT, action=Action.EXIT),
]

# Configure risk management
risk_config = TradingParameters(
    size=0.1,  # Use 10% of available capital per trade
    take_profit=0.05,  # Take profit at 5% price increase
    stop_loss=0.03,  # Cut losses at 3% price decrease
)

# Run the backtest
trades = process_signals(
    signals=signals,
    parameters=risk_config,
    fluctuations=fluctuations,
)

# Analyze results
print(f"Number of trades : {len(trades)}")
print(f"Total profit : {sum([trade.total_profit for trade in trades])}")
print(f"Win rate : {sum([1 for trade in trades if trade.total_profit > 0]) / len(trades)}")

Advanced Usage

Creating a Custom Strategy

You can define custom trading strategies by creating signals based on technical indicators or other market conditions:

from ptahlmud.backtesting.models.signal import Signal, Side, Action


def moving_average_strategy(fluctuations: Fluctuations, fast_period: int, slow_period: int) -> list[Signal]:
    """Moving average crossover strategy."""
    # Calculate moving averages (simplified example)
    fast_ma = fluctuations.dataframe["close"].rolling(fast_period).mean()
    slow_ma = fluctuations.dataframe["close"].rolling(slow_period).mean()

    ma_crossover = (fast_ma < slow_ma) & (fast_ma.shift(1) > slow_ma.shift(1))
    crossover_df = fluctuations.dataframe[ma_crossover]

    signals: list[Signal] = []
    for _, row in crossover_df.iterrows():
        signals.append(Signal(
            date=row["close_time"],
            action=Action.ENTER,
            side=Side.LONG
        ))

    return signals


signals = moving_average_strategy(fluctuations, fast_period=10, slow_period=30)
trades = process_signals(
    signals=signals,
    risk_config=risk_config,
    fluctuations=fluctuations,
)

Development

It is recommended to work in a virtual environment, you can install pyenv with python >= 3.11.

pyenv install 3.12.5
pyenv virtualenv 3.12.5 ptahlmud
pyenv activate ptahlmud

# Clone the repository
git clone https://github.com/yourusername/ptahlmud.git
cd ptahlmud

Setup environment

make setup

Run tests

make test

Run code quality checks

make check

Contributing

Contributions are welcome! You can open an issue, submit a pull-request or simply chat with me. I'm always pleased to discuss design, performance or technical stuff.