Hyperoptimizing Indicator Comparisons for Entry and Exit Signals
Code Freqtrade
Summary
This Freqtrade hyperoptimization module searches for rule-based long entries and exits using a catalog of price, volume, volatility, trend, and momentum series. For each rule, the search space selects an indicator, a second indicator or comparison target, a numeric threshold, and an operator. Operators include ordinary greater-than, less-than, and approximate equality checks, crossings, and comparisons against integer or real values. The generated entry and exit functions apply the selected condition to the market data frame.
The indicator list includes common measures such as RSI, moving averages, MACD, Bollinger-related fields, and volume indicators. The code defines separate parameter spaces for buy and sell rules and connects those spaces to Freqtrade’s hyperopt interface. It is a configurable search template, not evidence that any discovered rule is profitable. The document gives no market, timeframe, data split, transaction cost assumptions, or optimization results; users would need to evaluate overfitting, indicator warm-up behavior, and realistic execution separately.
Key ideas
- The module searches among technical indicator fields and comparison operators to form long entry and exit conditions.
- Rules can compare two indicator series or compare one series with an integer or real threshold.
- Cross-above and cross-below operators allow event-based conditions in addition to level comparisons.
- Buy and sell conditions are optimized through separate parameter spaces.
- The code provides no performance evidence or safeguards against overfitting.
Tags
Full text
# GodStraHo.py
```py
# GodStra Strategy Hyperopt
# Author: @Mablue (Masoud Azizi)
# github: https://github.com/mablue/
# IMPORTANT: INSTALL TA BEFORE RUN:
# :~$ pip install ta
# freqtrade hyperopt --hyperopt GodStraHo --hyperopt-loss SharpeHyperOptLossDaily --spaces all --strategy GodStra --config config.json -e 100
# --- Do not remove these libs ---
from functools import reduce
from typing import Any, Callable, Dict, List
import numpy as np # noqa
import pandas as pd # noqa
from pandas import DataFrame
from skopt.space import Categorical, Dimension, Integer, Real # noqa
from freqtrade.optimize.hyperopt_interface import IHyperOpt
# --------------------------------
# Add your lib to import here
# import talib.abstract as ta # noqa
from ta import add_all_ta_features
from ta.utils import dropna
import freqtrade.vendor.qtpylib.indicators as qtpylib
# this is your trading strategy DNA Size
# you can change it and see the results...
DNA_SIZE = 1
GodGenes = ["open", "high", "low", "close", "volume", "volume_adi", "volume_obv",
"volume_cmf", "volume_fi", "volume_mfi", "volume_em", "volume_sma_em", "volume_vpt",
"volume_nvi", "volume_vwap", "volatility_atr", "volatility_bbm", "volatility_bbh",
"volatility_bbl", "volatility_bbw", "volatility_bbp", "volatility_bbhi",
"volatility_bbli", "volatility_kcc", "volatility_kch", "volatility_kcl",
"volatility_kcw", "volatility_kcp", "volatility_kchi", "volatility_kcli",
"volatility_dcl", "volatility_dch", "volatility_dcm", "volatility_dcw",
"volatility_dcp", "volatility_ui", "trend_macd", "trend_macd_signal",
"trend_macd_diff", "trend_sma_fast", "trend_sma_slow", "trend_ema_fast",
"trend_ema_slow", "trend_adx", "trend_adx_pos", "trend_adx_neg", "trend_vortex_ind_pos",
"trend_vortex_ind_neg", "trend_vortex_ind_diff", "trend_trix",
"trend_mass_index", "trend_cci", "trend_dpo", "trend_kst",
"trend_kst_sig", "trend_kst_diff", "trend_ichimoku_conv",
"trend_ichimoku_base", "trend_ichimoku_a", "trend_ichimoku_b",
"trend_visual_ichimoku_a", "trend_visual_ichimoku_b", "trend_aroon_up",
"trend_aroon_down", "trend_aroon_ind", "trend_psar_up", "trend_psar_down",
"trend_psar_up_indicator", "trend_psar_down_indicator", "trend_stc",
"momentum_rsi", "momentum_stoch_rsi", "momentum_stoch_rsi_k",
"momentum_stoch_rsi_d", "momentum_tsi", "momentum_uo", "momentum_stoch",
"momentum_stoch_signal", "momentum_wr", "momentum_ao", "momentum_kama",
"momentum_roc", "momentum_ppo", "momentum_ppo_signal", "momentum_ppo_hist",
"others_dr", "others_dlr", "others_cr"]
class GodStraHo(IHyperOpt):
@staticmethod
def indicator_space() -> List[Dimension]:
"""
Define your Hyperopt space for searching buy strategy parameters.
"""
gene = list()
for i in range(DNA_SIZE):
gene.append(Categorical(GodGenes, name=f'buy-indicator-{i}'))
gene.append(Categorical(GodGenes, name=f'buy-cross-{i}'))
gene.append(Integer(-1, 101, name=f'buy-int-{i}'))
gene.append(Real(-1.1, 1.1, name=f'buy-real-{i}'))
# Operations
# CA: Crossed Above, CB: Crossed Below,
# I: Integer, R: Real, D: Disabled
gene.append(Categorical(["D", ">", "<", "=", "CA", "CB",
">I", "=I", "<I", ">R", "=R", "<R"], name=f'buy-oper-{i}'))
return gene
@staticmethod
def buy_strategy_generator(params: Dict[str, Any]) -> Callable:
"""
Define the buy strategy parameters to be used by Hyperopt.
"""
def populate_entry_trend(dataframe: DataFrame, metadata: dict) -> DataFrame:
"""
Buy strategy Hyperopt will build and use.
"""
conditions = []
# GUARDS AND TRENDS
for i in range(DNA_SIZE):
OPR = params[f'buy-oper-{i}']
IND = params[f'buy-indicator-{i}']
CRS = params[f'buy-cross-{i}']
INT = params[f'buy-int-{i}']
REAL = params[f'buy-real-{i}']
DFIND = dataframe[IND]
DFCRS = dataframe[CRS]
if OPR == ">":
conditions.append(DFIND > DFCRS)
elif OPR == "=":
conditions.append(np.isclose(DFIND, DFCRS))
elif OPR == "<":
conditions.append(DFIND < DFCRS)
elif OPR == "CA":
conditions.append(qtpylib.crossed_above(DFIND, DFCRS))
elif OPR == "CB":
conditions.append(qtpylib.crossed_below(DFIND, DFCRS))
elif OPR == ">I":
conditions.append(DFIND > INT)
elif OPR == "=I":
conditions.append(DFIND == INT)
elif OPR == "<I":
conditions.append(DFIND < INT)
elif OPR == ">R":
conditions.append(DFIND > REAL)
elif OPR == "=R":
conditions.append(np.isclose(DFIND, REAL))
elif OPR == "<R":
conditions.append(DFIND < REAL)
if conditions:
dataframe.loc[
reduce(lambda x, y: x & y, conditions),
'enter_long'] = 1
return dataframe
return populate_entry_trend
@ staticmethod
def sell_indicator_space() -> List[Dimension]:
"""
Define your Hyperopt space for searching sell strategy parameters.
"""
gene = list()
for i in range(DNA_SIZE):
gene.append(Categorical(GodGenes, name=f'sell-indicator-{i}'))
gene.append(Categorical(GodGenes, name=f'sell-cross-{i}'))
gene.append(Integer(-1, 101, name=f'sell-int-{i}'))
gene.append(Real(-0.01, 1.01, name=f'sell-real-{i}'))
# Operations
# CA: Crossed Above, CB: Crossed Below,
# I: Integer, R: Real, D: Disabled
gene.append(Categorical(["D", ">", "<", "=", "CA", "CB",
">I", "=I", "<I", ">R", "=R", "<R"], name=f'sell-oper-{i}'))
return gene
@ staticmethod
def sell_strategy_generator(params: Dict[str, Any]) -> Callable:
"""
Define the sell strategy parameters to be used by Hyperopt.
"""
def populate_exit_trend(dataframe: DataFrame, metadata: dict) -> DataFrame:
"""
Sell strategy Hyperopt will build and use.
"""
conditions = []
# GUARDS AND TRENDS
for i in range(DNA_SIZE):
OPR = params[f'sell-oper-{i}']
IND = params[f'sell-indicator-{i}']
CRS = params[f'sell-cross-{i}']
INT = params[f'sell-int-{i}']
REAL = params[f'sell-real-{i}']
DFIND = dataframe[IND]
DFCRS = dataframe[CRS]
if OPR == ">":
conditions.append(DFIND > DFCRS)
elif OPR == "=":
conditions.append(np.isclose(DFIND, DFCRS))
elif OPR == "<":
conditions.append(DFIND < DFCRS)
elif OPR == "CA":
conditions.append(qtpylib.crossed_above(DFIND, DFCRS))
elif OPR == "CB":
conditions.append(qtpylib.crossed_below(DFIND, DFCRS))
elif OPR == ">I":
conditions.append(DFIND > INT)
elif OPR == "=I":
conditions.append(DFIND == INT)
elif OPR == "<I":
conditions.append(DFIND < INT)
elif OPR == ">R":
conditions.append(DFIND > REAL)
elif OPR == "=R":
conditions.append(np.isclose(DFIND, REAL))
elif OPR == "<R":
conditions.append(DFIND < REAL)
if conditions:
dataframe.loc[
reduce(lambda x, y: x & y, conditions),
'exit_long']=1
return dataframe
return populate_exit_trend
```Shown in full with attribution under the source's licence. Licence: GPL-3.0
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.