Algorithmic trading has become increasingly popular among individual traders and financial institutions. Tools like freqtrade, an open-source cryptocurrency trading bot, allow traders to automate the implementation of trading strategies across various markets and exchanges. In this article, we will explore how to develop custom trading strategies using freqtrade.
Setting Up Freqtrade
To get started with freqtrade, you'll need to have Python installed on your system. Freqtrade requires Python version 3.6 or higher. You should also have a basic understanding of both Python and trading concepts.
$ git clone https://github.com/freqtrade/freqtrade.git
$ cd freqtrade
$ python3 -m venv .env
$ source .env/bin/activate
$ pip install -r requirements.txt
The above steps help you to clone the freqtrade repository, create a virtual environment, and install the necessary dependencies.
Understanding Freqtrade's Architecture
Freqtrade's architecture revolves around the separation of strategy logic from the core bot functionality. This approach allows you to focus on building your strategy without worrying about the other operational aspects of the bot.
Creating a Custom Strategy
Your trading strategy is defined by a Python class that specifies how and when orders should be placed and executed. Let's define a simple example strategy:
from freqtrade.strategy import IStrategy
from pandas import DataFrame
class SimpleStrategy(IStrategy):
def __init__(self):
super().__init__()
# Define the minimal ROI values
minimal_roi = {
"0": 0.1,
"10": 0.02
}
# Define stop loss
stoploss = -0.10
# Define trailing stop
trailing_stop = True
trailing_stop_positive = 0.01
def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe['ma'] = dataframe['close'].rolling(window=10).mean() # Simple Moving Average
return dataframe
def populate_entry_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe.loc[(dataframe['close'] > dataframe['ma']), 'enter_long'] = 1
return dataframe
def populate_exit_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe.loc[(dataframe['close'] < dataframe['ma']), 'exit_long'] = 1
return dataframe
This simple strategy uses a simple moving average (SMA) to determine entry and exit points. The strategy buys when the price goes above the SMA and sells when it goes below.
Running Your Strategy
To run your custom strategy with freqtrade, you'll need to adjust the bot's configuration file to specify which strategy to use. Here is how you can do it:
{
"strategy": "SimpleStrategy",
"start_balance": 1000,
"fee": 0.1,
"exchange": {
"name": "binance",
"key": "",
"secret": "",
"ccxt_config": {},
"ccxt_async_config": {},
"pair_whitelist": [
"BTC/USDT"
]
}
}
After updating the configuration, activate your virtual environment if not already activated and launch the bot:
$ freqtrade tradeBacktesting Strategies
Before running a strategy live, it's crucial to backtest it using historical data to verify its profitability. Freqtrade provides tools to perform thorough backtests.
$ freqtrade backtesting --strategy SimpleStrategy --refresh-pairs-cachedOptimizing the Strategy
Optimization involves tweaking the strategy parameters to achieve better returns. Freqtrade includes optimization capabilities that perform parameter sweeps or genetic optimizations to find the most profitable configuration.
$ freqtrade hyperopt --strategy SimpleStrategyBy fine-tuning your strategy's parameters, you may see substantial improvements in strategy performance while assessing risk factors associated with different configurations.
Final Thoughts
Developing custom trading strategies with freqtrade opens the door to automating your trading, allowing you to explore both simple and complex algorithmic structures. By continuously refining and backtesting your strategies, you can discover new market opportunities and techniques suitable for your risk appetite and investment goals. As with any trading software, it's important to monitor your bots regularly to ensure everything runs smoothly and adjustments can be made as market conditions evolve.
