Lucien EMA strategy

data/leaderboard.json

@@ -181,43 +181,23 @@
             }
         },
         {
-            "author_name": "Example Author",
-            "strategy_name": "Example Strategy",
-            "description": "An example strategy",
-            "development_metrics": {
-                "sharpe": 0.0,
-                "total_return": 0.0,
-                "max_drawdown": 0.0,
-                "n_trades": 0,
-                "win_rate": 0.0
-            },
-            "holdout_metrics": {
-                "sharpe": 0.0,
-                "total_return": 0.0,
-                "max_drawdown": 0.0,
-                "n_trades": 0,
-                "win_rate": 0.0
-            },
-            "last_updated": "2024-01-01T00:00:00Z"
-        },
-        {
-            "author_name": "YOUR_NAME",
-            "strategy_name": "YOUR_STRATEGY_NAME",
-            "description": "DESCRIBE YOUR STRATEGY HERE",
-            "last_updated": "2025-05-17T08:23:59.481326Z",
+            "author_name": "Lucien",
+            "strategy_name": "EMA Crossover + Adaptive ATR Filter",
+            "description": "Buy when fast EMA > slow EMA and ATR% > rolling quantile threshold; sell on reversal or low adaptive volatility.",
+            "last_updated": "2025-05-17T12:47:24.735129Z",
             "development_metrics": {
-                "sharpe": 0.0,
-                "total_return": 0.0,
-                "max_drawdown": 0.0,
-                "n_trades": 0,
-                "win_rate": 0.0
+                "sharpe": 1.217803722952385,
+                "total_return": 36432.64411560105,
+                "max_drawdown": 0.8861273181842388,
+                "n_trades": 2787,
+                "win_rate": 0.5331898098313599
             },
             "holdout_metrics": {
-                "sharpe": 0.0,
-                "total_return": 0.0,
-                "max_drawdown": 0.0,
-                "n_trades": 0,
-                "win_rate": 0.0
+                "sharpe": 1.2721023608824928,
+                "total_return": 0.11707790357354719,
+                "max_drawdown": 0.16707933054084675,
+                "n_trades": 76,
+                "win_rate": 0.5263157894736842
             }
         }
     ]

src/strategies/lucien_ema-v1.py

@@ -0,0 +1,62 @@
+"""
+EMA Crossover + Adaptive ATR Filter strategy for Bitcoin.
+Buy when fast EMA > slow EMA and ATR% exceeds its rolling quantile threshold.
+Sell when EMA crosses down or ATR% falls below that adaptive threshold.
+"""
+
+import pandas as pd
+from src.core.strategy import Strategy  # adjust import as needed
+
+class EMACrossoverWithAdaptiveATRFilterStrategy(Strategy):
+    def __init__(self,
+                 initial_capital=10000,
+                 fast=30,
+                 slow=120,
+                 atr_period=14,
+                 atr_quantile_window=100,
+                 atr_quantile=0.5):
+        super().__init__(
+            initial_capital=initial_capital,
+            author_name="Lucien",
+            strategy_name="EMA Crossover + Adaptive ATR Filter",
+            description=(
+                "Buy when fast EMA > slow EMA and ATR% > rolling "
+                "quantile threshold; sell on reversal or low adaptive volatility."
+            )
+        )
+        self.fast = fast
+        self.slow = slow
+        self.atr_period = atr_period
+        self.atr_quantile_window = atr_quantile_window
+        self.atr_quantile = atr_quantile
+
+    def get_signals(self, df: pd.DataFrame) -> pd.Series:
+        signals = pd.Series('hold', index=df.index)
+
+        # 1. Compute EMAs
+        ema_fast = df['close'].ewm(span=self.fast, adjust=False).mean()
+        ema_slow = df['close'].ewm(span=self.slow, adjust=False).mean()
+
+        # 2. Compute ATR% (using close-only TR approximation)
+        prev_close = df['close'].shift(1)
+        tr = (df['close'] - prev_close).abs()
+        atr = tr.rolling(window=self.atr_period).mean()
+        atr_pct = atr / df['close']
+
+        # 3. Build adaptive threshold: rolling quantile of ATR%
+        atr_threshold = atr_pct.rolling(window=self.atr_quantile_window) \
+                               .quantile(self.atr_quantile)
+
+        # 4. Entry / Exit conditions
+        buy_cond = (ema_fast > ema_slow) & (atr_pct > atr_threshold)
+        sell_cond = (ema_fast < ema_slow) | (atr_pct < atr_threshold)
+
+        signals[buy_cond] = 'buy'
+        signals[sell_cond] = 'sell'
+
+        # 5. Warm-up / prevent lookahead
+        warmup = max(self.slow, self.atr_period, self.atr_quantile_window)
+        signals.iloc[:warmup] = 'hold'
+
+        # 6. Shift by one to avoid lookahead bias
+        return signals.shift(1).fillna('hold')