harry/ForexBot2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Round-2 weight optimisation — reduce EMA trend, boost stochastic

Browse Source 
GPU sweep (36 combos, RTX 4070 SUPER) on H4 2022-2026 relative to the
round-1 baseline revealed EMA trend was over-weighted at 3.6. Halving
it and nudging stochastic up yields cleaner, higher-quality signals.

Weight changes (backtest_gpu.py _BASE_W + engine.py):
  ema_trend:  3.6 -> 1.8  (over-weighted; halving tightens signal filter)
  stochastic: 1.6 -> 2.4  (x1.5; improved WR to 40.4% in sweep)
  macd:       3.3          (unchanged — dominant signal, kept as anchor)
  adx/ichimoku/rsi/bollinger: unchanged

Combined result vs round-1 (H4 local data 2022-2026):
  Trades:      160 -> 73   (-87 — filters out marginal setups)
  Win Rate:   38.1% -> 43.8%  (+5.7pp)
  Profit Factor: 1.19 -> 1.52
  Return:    +21.30% -> +23.86%  (+2.57pp)
  Max DD:     -8.52% -> -9.32%
  Sharpe:      0.47  ->  0.71   (+51%)
  Sortino:     1.91  ->  3.06   (+60%)

Fewer trades at much higher quality. The lower EMA weight means only
strong, multi-indicator confluences clear the grade threshold — the
73 remaining trades have 43.8% WR vs 38.1% and expectancy nearly 3x.
This commit is contained in:
Harwinder
2026-05-07 22:24:48 -04:00
parent e57f4a1ecd
commit a55aca21b1
2 changed files with 14 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backtest_gpu.py
+12 -10
View File
@@ -408,19 +408,21 @@ class VectorIndicators:
# ══════════════════════════════════════════════════════════════════
# Per-indicator base weights
# Optimised 2026-05-07 via 36-combo sweep on H4 2022-2026 local data:
# ema_trend ×3, macd ×3, stochastic ×2, adx ×0.5, ichimoku ×0.5, rsi=0, bollinger=0
# Combined result: +21.30% / -8.52% DD / Sharpe 0.47 vs baseline +7.12% / -9.05% DD / 0.23
# Round 1 optimisation 2026-05-07: ema_trend=3.6, macd=3.3, stoch=1.6, adx=0.5, ichi=0.55
# → +21.30% / -8.52% DD / Sharpe 0.47
# Round 2 optimisation 2026-05-08 (GPU sweep relative to round-1 baseline):
# ema_trend ×0.5 was best single move: +32.68% / -6.15% DD / Sharpe 0.74
# stochastic ×1.5 added modest improvement; macd kept unchanged
# Original weights (engine.py defaults, pre-optimisation):
# ema_trend=1.2, rsi=0.3, macd=1.1, bollinger=0.9, stochastic=0.8, adx=1.0, ichimoku=1.1
_BASE_W = {
"ema_trend": 3.6,
"macd": 3.3,
"stochastic": 1.6,
"adx": 0.5,
"ichimoku": 0.55,
"rsi": 0.0, # ablated — net drag on H4 2022-2026
"bollinger": 0.0, # ablated — counter-trend noise on trending pair
"ema_trend": 1.8, # was 3.6 — over-weighted; halving lifts return +11pp, tightens DD
"macd": 3.3, # unchanged — dominant signal
"stochastic": 2.4, # was 1.6 — ×1.5 improved WR to 40.4%
"adx": 0.5, # unchanged
"ichimoku": 0.55, # unchanged
"rsi": 0.0, # ablated — confirmed: all multipliers give identical result
"bollinger": 0.0, # ablated — confirmed: all multipliers give identical result
}
# Regime-based weight multipliers — matches engine._compute_composite
engine.py
+2 -2
View File
@@ -171,7 +171,7 @@ class TechnicalEngine:
result.indicators["ema_trend"] = IndicatorSignal(
name="EMA Trend", value=normalized,
signal=signal, confidence=conf,
description=desc, weight=3.6,
description=desc, weight=1.8,
)
if t is not None:
result.key_levels["ema_trend"] = float(t)
@@ -423,7 +423,7 @@ class TechnicalEngine:
result.indicators["stochastic"] = IndicatorSignal(
name="Stochastic", value=k_val, signal=signal,
confidence=conf, description=desc, weight=1.6,
confidence=conf, description=desc, weight=2.4,
)
def _compute_adx(self, df: pd.DataFrame, result: TechnicalAnalysis):