From: Jérôme Benoit <jerome.benoit@piment-noir.org>
Date: Mon, 24 Mar 2025 17:18:20 +0000 (+0100)
Subject: fix(qav3): fix zero lag moving average implementation
X-Git-Url: https://git.piment-noir.org/?a=commitdiff_plain;h=5a8ae7a248e93411d0d4c7946d22354cee810ce1;p=freqai-strategies.git

fix(qav3): fix zero lag moving average implementation

Signed-off-by: Jérôme Benoit <jerome.benoit@piment-noir.org>
---

diff --git a/quickadapter/user_data/strategies/QuickAdapterV3.py b/quickadapter/user_data/strategies/QuickAdapterV3.py
index 4c343e6..8a4acb2 100644
--- a/quickadapter/user_data/strategies/QuickAdapterV3.py
+++ b/quickadapter/user_data/strategies/QuickAdapterV3.py
@@ -259,7 +259,7 @@ class QuickAdapterV3(IStrategy):
             non_zero_range(dataframe["high"], dataframe["low"])
         )
         dataframe["jaw"], dataframe["teeth"], dataframe["lips"] = alligator(
-            dataframe, zero_lag=True
+            dataframe, pricemode="median", mamode="sma", zero_lag=True
         )
         dataframe["%-dist_to_jaw"] = get_distance(dataframe["close"], dataframe["jaw"])
         dataframe["%-dist_to_teeth"] = get_distance(
diff --git a/quickadapter/user_data/strategies/Utils.py b/quickadapter/user_data/strategies/Utils.py
index 0486385..3d9e076 100644
--- a/quickadapter/user_data/strategies/Utils.py
+++ b/quickadapter/user_data/strategies/Utils.py
@@ -1,10 +1,8 @@
 import numpy as np
 import pandas as pd
-import pandas_ta as pta
 import talib.abstract as ta
 from scipy.signal import convolve
 from scipy.signal.windows import gaussian
-from sys import float_info
 from technical import qtpylib
 
 
@@ -15,8 +13,7 @@ def get_distance(p1: pd.Series | float, p2: pd.Series | float) -> pd.Series | fl
 def non_zero_range(s1: pd.Series, s2: pd.Series) -> pd.Series:
     """Returns the difference of two series and adds epsilon to any zero values."""
     diff = s1 - s2
-    if diff.eq(0).any().any():
-        diff += float_info.epsilon
+    diff = diff.mask(diff == 0, other=diff + np.finfo(float).eps)
     return diff
 
 
@@ -124,7 +121,12 @@ def vwapb(dataframe: pd.DataFrame, window=20, num_of_std=1) -> tuple:
     return vwap_low, vwap, vwap_high
 
 
-def get_ma_fn(mamode: str, zero_lag=False) -> callable:
+def zero_lag_series(series: pd.Series, timeperiod: int) -> pd.Series:
+    lag = int(0.5 * (timeperiod - 1))
+    return 2 * series - series.shift(lag)
+
+
+def get_ma_fn(mamode: str) -> callable:
     mamodes: dict = {
         "sma": ta.SMA,
         "ema": ta.EMA,
@@ -135,13 +137,7 @@ def get_ma_fn(mamode: str, zero_lag=False) -> callable:
         "kama": ta.KAMA,
         "t3": ta.T3,
     }
-    if zero_lag:
-        ma_fn = lambda series, timeperiod: pta.zlma(
-            series, length=timeperiod, mamode=mamode
-        )
-    else:
-        ma_fn = mamodes.get(mamode, mamodes["sma"])
-    return ma_fn
+    return mamodes.get(mamode, mamodes["sma"])
 
 
 def fractal_dimension(
@@ -224,11 +220,12 @@ def smma(
     if len(series) < period:
         return pd.Series(index=series.index, dtype=float)
 
+    if zero_lag:
+        series = zero_lag_series(series, timeperiod=period)
     smma = series.copy()
     smma[: period - 1] = np.nan
-    ma_fn = get_ma_fn(mamode, zero_lag=zero_lag)
     smma.iloc[period - 1] = pd.Series(
-        ma_fn(series.iloc[:period], timeperiod=period)
+        get_ma_fn(mamode)(series.iloc[:period], timeperiod=period)
     ).iloc[-1]
 
     for i in range(period, len(series)):
@@ -261,10 +258,16 @@ def ewo(
     """
     Calculate the Elliott Wave Oscillator (EWO) using two moving averages.
     """
+    ma_fn = get_ma_fn(mamode)
     price_series = get_price_fn(pricemode)(dataframe)
-    ma_fn = get_ma_fn(mamode, zero_lag=zero_lag)
-    ma1 = ma_fn(price_series, timeperiod=ma1_length)
-    ma2 = ma_fn(price_series, timeperiod=ma2_length)
+    if zero_lag:
+        price_series_ma1_length = zero_lag_series(price_series, timeperiod=ma1_length)
+        price_series_ma2_length = zero_lag_series(price_series, timeperiod=ma2_length)
+        ma1 = ma_fn(price_series_ma1_length, timeperiod=ma1_length)
+        ma2 = ma_fn(price_series_ma2_length, timeperiod=ma2_length)
+    else:
+        ma1 = ma_fn(price_series, timeperiod=ma1_length)
+        ma2 = ma_fn(price_series, timeperiod=ma2_length)
     madiff = ma1 - ma2
     if normalize:
         madiff = (madiff / price_series) * 100