stable/freqtrade/optimize/hyperopt_loss_calmar_daily.py

80 lines
2.3 KiB
Python

"""
CalmarHyperOptLossDaily
This module defines the alternative HyperOptLoss class which can be used for
Hyperoptimization.
"""
from datetime import datetime
from math import sqrt as msqrt
from typing import Any, Dict
from pandas import DataFrame, date_range
from freqtrade.optimize.hyperopt import IHyperOptLoss
class CalmarHyperOptLossDaily(IHyperOptLoss):
"""
Defines the loss function for hyperopt.
This implementation uses the Calmar Ratio calculation.
"""
@staticmethod
def hyperopt_loss_function(
results: DataFrame,
trade_count: int,
min_date: datetime,
max_date: datetime,
backtest_stats: Dict[str, Any],
*args,
**kwargs
) -> float:
"""
Objective function, returns smaller number for more optimal results.
Uses Calmar Ratio calculation.
"""
resample_freq = "1D"
slippage_per_trade_ratio = 0.0005
days_in_year = 365
# create the index within the min_date and end max_date
t_index = date_range(
start=min_date, end=max_date, freq=resample_freq, normalize=True
)
# apply slippage per trade to profit_total
results.loc[:, "profit_ratio_after_slippage"] = (
results["profit_ratio"] - slippage_per_trade_ratio
)
sum_daily = (
results.resample(resample_freq, on="close_date")
.agg({"profit_ratio_after_slippage": sum})
.reindex(t_index)
.fillna(0)
)
total_profit = sum_daily["profit_ratio_after_slippage"]
expected_returns_mean = total_profit.mean() * 100
# calculate max drawdown
try:
high_val = total_profit.max()
low_val = total_profit.min()
max_drawdown = (high_val - low_val) / high_val
except (ValueError, ZeroDivisionError):
max_drawdown = 0
if max_drawdown != 0:
calmar_ratio = expected_returns_mean / max_drawdown * msqrt(days_in_year)
else:
# Define high (negative) calmar ratio to be clear that this is NOT optimal.
calmar_ratio = -20.0
# print(t_index, sum_daily, total_profit)
# print(expected_returns_mean, max_drawdown, calmar_ratio)
return -calmar_ratio