Merge pull request #5607 from TreborNamor/develop

a new hyperopt loss created that uses calmar ratio

docs/hyperopt.md

@@ -116,7 +116,7 @@ optional arguments:
                         ShortTradeDurHyperOptLoss, OnlyProfitHyperOptLoss,
                         SharpeHyperOptLoss, SharpeHyperOptLossDaily,
                         SortinoHyperOptLoss, SortinoHyperOptLossDaily,
-                        MaxDrawDownHyperOptLoss
+                        CalmarHyperOptLoss, MaxDrawDownHyperOptLoss
   --disable-param-export
                         Disable automatic hyperopt parameter export.
   --ignore-missing-spaces, --ignore-unparameterized-spaces
@@ -524,6 +524,7 @@ Currently, the following loss functions are builtin:
 * `SortinoHyperOptLoss` - optimizes Sortino Ratio calculated on trade returns relative to **downside** standard deviation.
 * `SortinoHyperOptLossDaily` - optimizes Sortino Ratio calculated on **daily** trade returns relative to **downside** standard deviation.
 * `MaxDrawDownHyperOptLoss` - Optimizes Maximum drawdown.
+* `CalmarHyperOptLoss` - Optimizes Calmar Ratio calculated on trade returns relative to max drawdown.
 
 Creation of a custom loss function is covered in the [Advanced Hyperopt](advanced-hyperopt.md) part of the documentation.
 

freqtrade/constants.py

@@ -25,6 +25,7 @@ ORDERTIF_POSSIBILITIES = ['gtc', 'fok', 'ioc']
 HYPEROPT_LOSS_BUILTIN = ['ShortTradeDurHyperOptLoss', 'OnlyProfitHyperOptLoss',
                          'SharpeHyperOptLoss', 'SharpeHyperOptLossDaily',
                          'SortinoHyperOptLoss', 'SortinoHyperOptLossDaily',
+                         'CalmarHyperOptLoss',
                          'MaxDrawDownHyperOptLoss']
 AVAILABLE_PAIRLISTS = ['StaticPairList', 'VolumePairList',
                        'AgeFilter', 'OffsetFilter', 'PerformanceFilter',
@@ -53,7 +54,6 @@ ENV_VAR_PREFIX = 'FREQTRADE__'
 
 NON_OPEN_EXCHANGE_STATES = ('cancelled', 'canceled', 'closed', 'expired')
 
-
 # Define decimals per coin for outputs
 # Only used for outputs.
 DECIMAL_PER_COIN_FALLBACK = 3  # Should be low to avoid listing all possible FIAT's
@@ -67,7 +67,6 @@ DUST_PER_COIN = {
     'ETH': 0.01
 }
 
-
 # Source files with destination directories within user-directory
 USER_DATA_FILES = {
     'sample_strategy.py': USERPATH_STRATEGIES,

freqtrade/optimize/hyperopt_loss_calmar.py

@@ -0,0 +1,64 @@
+"""
+CalmarHyperOptLoss
+
+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
+
+from freqtrade.data.btanalysis import calculate_max_drawdown
+from freqtrade.optimize.hyperopt import IHyperOptLoss
+
+
+class CalmarHyperOptLoss(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,
+        config: Dict,
+        processed: Dict[str, DataFrame],
+        backtest_stats: Dict[str, Any],
+        *args,
+        **kwargs
+    ) -> float:
+        """
+        Objective function, returns smaller number for more optimal results.
+
+        Uses Calmar Ratio calculation.
+        """
+        total_profit = backtest_stats["profit_total"]
+        days_period = (max_date - min_date).days
+
+        # adding slippage of 0.1% per trade
+        total_profit = total_profit - 0.0005
+        expected_returns_mean = total_profit.sum() / days_period * 100
+
+        # calculate max drawdown
+        try:
+            _, _, _, high_val, low_val = calculate_max_drawdown(
+                results, value_col="profit_abs"
+            )
+            max_drawdown = (high_val - low_val) / high_val
+        except ValueError:
+            max_drawdown = 0
+
+        if max_drawdown != 0:
+            calmar_ratio = expected_returns_mean / max_drawdown * msqrt(365)
+        else:
+            # Define high (negative) calmar ratio to be clear that this is NOT optimal.
+            calmar_ratio = -20.0
+
+        # print(expected_returns_mean, max_drawdown, calmar_ratio)
+        return -calmar_ratio

tests/optimize/test_hyperoptloss.py

@@ -85,6 +85,8 @@ def test_loss_calculation_has_limited_profit(hyperopt_conf, hyperopt_results) ->
     "SharpeHyperOptLoss",
     "SharpeHyperOptLossDaily",
     "MaxDrawDownHyperOptLoss",
+    "CalmarHyperOptLoss",
+
 ])
 def test_loss_functions_better_profits(default_conf, hyperopt_results, lossfunction) -> None:
     results_over = hyperopt_results.copy()
@@ -96,11 +98,32 @@ def test_loss_functions_better_profits(default_conf, hyperopt_results, lossfunct
 
     default_conf.update({'hyperopt_loss': lossfunction})
     hl = HyperOptLossResolver.load_hyperoptloss(default_conf)
-    correct = hl.hyperopt_loss_function(hyperopt_results, len(hyperopt_results),
+    correct = hl.hyperopt_loss_function(
-                                        datetime(2019, 1, 1), datetime(2019, 5, 1))
+        hyperopt_results,
-    over = hl.hyperopt_loss_function(results_over, len(results_over),
+        trade_count=len(hyperopt_results),
-                                     datetime(2019, 1, 1), datetime(2019, 5, 1))
+        min_date=datetime(2019, 1, 1),
-    under = hl.hyperopt_loss_function(results_under, len(results_under),
+        max_date=datetime(2019, 5, 1),
-                                      datetime(2019, 1, 1), datetime(2019, 5, 1))
+        config=default_conf,
+        processed=None,
+        backtest_stats={'profit_total': hyperopt_results['profit_abs'].sum()}
+        )
+    over = hl.hyperopt_loss_function(
+        results_over,
+        trade_count=len(results_over),
+        min_date=datetime(2019, 1, 1),
+        max_date=datetime(2019, 5, 1),
+        config=default_conf,
+        processed=None,
+        backtest_stats={'profit_total': results_over['profit_abs'].sum()}
+    )
+    under = hl.hyperopt_loss_function(
+        results_under,
+        trade_count=len(results_under),
+        min_date=datetime(2019, 1, 1),
+        max_date=datetime(2019, 5, 1),
+        config=default_conf,
+        processed=None,
+        backtest_stats={'profit_total': results_under['profit_abs'].sum()}
+    )
     assert over < correct
     assert under > correct