`Condition` — Strategy Condition Module

This module defines the various condition classes used to create complex entry and exit logic in trading strategies. These conditions are categorized into logical, threshold-based, trend-following, Fibonacci retracement and parameterized (time/value-based) mechanisms.

All condition classes inherit from the base Condition class, which provides a common interface for evaluating conditions against time series data.

Abstract Condition class has the following methods:

class Condition(ABC):

    @abstractmethod
    def evaluate(self, data: pd.DataFrame) -> pd.Series:
        pass

    @abstractmethod
    def get_graphs(self, downloader: DownloadModule, df: pd.DataFrame) -> [TradingPlot]:
        pass

    @abstractmethod
    def to_string(self) -> str:
        pass

    @abstractmethod
    def to_dict(self) -> dict:
        pass

As a condition is evaluated over a time series, it returns a boolean series indicating whether the condition is satisfied at each point in time. These boolean series can be combined using logical conditions to create complex strategies and are stored in the pd.DataFrame.

Module Structure

1. Logical Conditions

Located in trading_strategy_tester/conditions/logical_conditions/

AND — Combines multiple conditions using logical AND
OR — Combines multiple conditions using logical OR

2. Threshold Conditions

Located in trading_strategy_tester/conditions/threshold_conditions/

CrossOverCondition — Condition when one series crosses over another
CrossUnderCondition — Condition when one series crosses under another
GreaterThanCondition — Checks if one series is greater than another
LessThanCondition — Checks if one series is less than another

3. Trend Conditions

Located in trading_strategy_tester/conditions/trend_conditions/

UptrendForXDaysCondition — Detects uptrend over X consecutive days
DowntrendForXDaysCondition — Detects downtrend over X consecutive days

4. Fibonacci Retracement Conditions

Located in trading_strategy_tester/conditions/fibonacci_retracement_levels_conditions/

UptrendFibRetracementLevelCondition — Uses Fibonacci levels during uptrend
DowntrendFibRetracementLevelCondition — Uses Fibonacci levels during downtrend

5. Parameterized Conditions

Located in trading_strategy_tester/conditions/parameterized_conditions/

AfterXDaysCondition — Triggers condition after X days of another condition
ChangeOfXPercentPerYDaysCondition — Checks percentage change over Y days
IntraIntervalChangeOfXPercentCondition — Percentage change within a single interval chosen in the strategy definition

Integration

Each condition class is typically used inside a Strategy object as a buy_condition or sell_condition. Conditions can be nested using logical conditions for complex rule-based strategies.

Condition — Strategy Condition Module