Metis

CFD trading simulation for back- and forward testing different trading strategies.

⭐ Please star this repository — it motivates a lot!

🚀 About

Stocks and CFD's are a great way to start trading and earn some money. But what strategy should I use to actually make money 💵. This is where this simulator comes in. You can use predefined strategies or implement your own strategy and test how well it would perform long-term on any CFD listed by yfinance.

📝 Getting started

Prerequisites

• Python >=3.13

Installation

Note

Note that the virtual environment always has to be active when trying to run simulations.

• Clone repository: git clone https://github.com/engag1ng/stock-simulation.git
• Create virtual environment: python -m venv .venv
• Activate virtual environment: .venv\Scripts\activate
• Install dependencies: pip install -r requirements.txt

Usage

1. Replace the parameters in MAIN.py, to your liking.
2. Run python MAIN.py

ℹ️ Documentation

Before adding new features, indicators or strategies, it is important to understand, what each one is intended for.

Features: provide the ability to calculate values (columns) of data, that help make informed decisions for trading or just help to calcuate other values.

Indicators: typically returns a value between -10 and 10 which indicates, if we should buy (positive) or sell (negative) and how much, where 0 indicates holding, 10 indicates going all-in bullish and -10 indicates all-in bearish.

Strategies: Combining features and indicators to create an algorithm, that makes trading decisions, like buying / selling and setting stop-loss/take-profit.

Add a New Feature

Location: features/

1. Create a new file in the features/ directory.
2. Define your feature function, for example:

def feature_daily_return(df):
    return df['Close'].pct_change().fillna(0)

Add a New Indicator

Location: indicators/

1. Create a new file or add to an existing one in the indicators/ directory.
2. Use the @register_indicator decorator to register it automatically:

from indicators import register_indicator

@register_indicator
def indicator_my_custom(df):
    return df['Close'].rolling(window=10).mean()

Add a New Strategy

Location: strategies/

1. Create a new file or add to an existing one in the strategies/ directory.
2. Subclass the Strategy base class and use the @Strategy.register decorator with a unique name:

from strategies.base import Strategy

@Strategy.register("my_custom_strategy")
class MyCustomStrategy(Strategy):
    def execute(self, df, owned_stocks, price, capital):
        # Your buy/sell logic here
        return 0.1, [price * 0.95, 0.1], [price * 1.05, 0.1]

⏱️ Roadmap

• Implement new strategies
• Add extensive documentation
• Improve simulation sequencing
• Improve UI

🤝 Contributions

Community contributions are very welcome. Please see below how you can contribute...

1. Find bugs and report them.
2. Make recommendations.
3. Make improvements to the code base.
4. Make improvements to the documentation.

📃 License

This project is GNU GPLv3 licensed. Please have a look at the LICENSE.md for more information.