Skip to content

Strategy

Strategy

Bases: ABC

Abstract base class for trading strategies in the backtesting framework.

This class defines the interface that all trading strategies must implement and provides core functionality for managing data sources, positions, and technical indicators. Strategies inherit from this class to implement their specific trading logic.

The Strategy class manages: - Multiple data sources and their associated broker positions - Technical indicator arrays using TimeNDArray - Initialization and iteration over market data - Abstract methods for strategy-specific logic

Example

class MyStrategy(Strategy):
... def init(self):
... # Initialize indicators and parameters
... self.fast_ma = None
...
... def next(self):
... # Trading logic for each time step
... if len(self.data['AAPL'].Close) > 20:
... fast = np.mean(self.data['AAPL'].Close[-10:])
... slow = np.mean(self.data['AAPL'].Close[-20:])
... if fast > slow:
... self.positions['AAPL'].buy(quantity=0.1)
... else:
... self.positions['AAPL'].sell(quantity=0.1)

Indicator(arr)

Create a time-aware indicator array from input data.

This method converts a numpy array into a TimeNDArray, which maintains time awareness for backtesting. The returned array can be used to store technical indicators that need to be progressively built during the backtest.

Parameters:

Name Type Description Default
arr NDArray

Input array containing indicator values.

 required

Returns:

Name Type Description
TimeNDArray

Time-aware array that can be used for indicators.
Note

This method is final and cannot be overridden by strategy classes. The indicator is automatically added to the strategy's indicators list and will be properly managed during backtest iteration.

Example

class MyStrategy(Strategy):
... def init(self):
... # Create empty indicator array
... self.sma = self.Indicator(np.zeros(len(self.data['AAPL'])))
...
... def next(self):
... # Update indicator
... if len(self.data['AAPL'].Close) >= 20:
... self.sma[-1] = np.mean(self.data['AAPL'].Close[-20:])

__init__()

Initialize the strategy with empty data structures.

This constructor sets up the core containers for data sources, positions, and indicators that the strategy will use.

add_data(source, symbol)

Add a data source and corresponding broker position to the strategy.

This method associates a data source (containing market data for a specific symbol) with a broker position for that same symbol. It's typically called during strategy initialization.

Parameters:

Name Type Description Default
source DataSource

DataSource object containing OHLCV data for the specified symbol.

 required
symbol str

Identifier for the symbol (e.g., 'AAPL', 'EURUSD').

 required
Note

This method is final and cannot be overridden by strategy classes. Each symbol can only have one data source and broker position.

Example

class MyStrategy(Strategy):
... def init(self):
... self.add_data(CSVDataSource("AAPL.csv"), "AAPL")
... def next(self):
... if self.data['AAPL'].CClose > self.data['AAPL'].Close[-1]:
... self.positions['AAPL'].buy(quantity=0.1)

init() abstractmethod

Initialize the strategy with indicators and parameters.

This method is called once at the beginning of the backtest before any data iteration begins. Strategies should use this method to: - Initialize technical indicators - Set strategy parameters - Prepare any pre-computed values - Setup any required data structures

Note

This method must be implemented by concrete strategy classes. It is called automatically by the backtester.

next() abstractmethod

Execute strategy logic for the current time step.

This method is called for each time step in the backtest data and should contain the main trading logic of the strategy. At the time this method is called, all data up to the current time step is available in the data sources and positions.

The method can access: - self.data: Dictionary of DataSource objects with current market data - self.positions: Dictionary of Broker objects for position management - self.indicators: List of TimeNDArray objects for technical analysis

Typical actions include: - Reading current market data (e.g., self.data['AAPL'].CClose) - Updating technical indicators - Making buy/sell decisions based on indicators - Managing position sizes - Setting stop loss or take profit levels

Note

This method must be implemented by concrete strategy classes. It is called automatically by the backtester for each time step.

Source code