Evaluate weights against next-period returns to avoid look-ahead

Weights formed from close[t] now earn the t→t+1 return: forward returns
are computed on the full market calendar before selecting signal dates,
so a sparse signal grid earns the next available return rather than the
next signal date, and the final signal date (no forward return) is
dropped. Signal pct_change uses fill_method=None so suspended names do
not inherit stale prices.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

pipeline/portfolio/research.py

@@ -35,7 +35,7 @@ def evaluate_portfolio(positions_df: pd.DataFrame, data_df: pd.DataFrame) -> dic
     close = data_df.pivot_table(
         index="date", columns="symbol_id", values="close", aggfunc="first"
     ).sort_index()
-    returns = close.pct_change()
+    returns = close.pct_change(fill_method=None)
 
     weights = positions_df.pivot_table(
         index="date", columns="symbol_id", values="target_weight", aggfunc="first"
@@ -43,22 +43,29 @@ def evaluate_portfolio(positions_df: pd.DataFrame, data_df: pd.DataFrame) -> dic
 
     common = weights.index.intersection(returns.index)
     weights = weights.loc[common]
-    returns = returns.loc[common]
+    # Compute forward returns on the full market calendar before selecting
+    # signal dates. This preserves next-period returns when the signal grid is
+    # sparser than the data grid.
+    fwd = returns.shift(-1).reindex(common)
 
     empty = {
         "cumulative_return": 0.0, "sharpe_annual": 0.0, "turnover_annual": 0.0,
         "max_drawdown": 0.0, "fitness": 0.0, "hit_rate": 0.0,
         "n_dates": len(common),
     }
-    if len(common) < 3:
+    if len(common) < 1:
+        empty["n_dates"] = 0
         return empty
 
     gross = weights.abs().sum(axis=1)
-    # Weights at t earn the return from t to t+1: shift returns back by one.
-    fwd = returns.shift(-1)
-    daily = (weights * fwd).sum(axis=1) / gross.replace(0.0, np.nan)
+    # Weights at t earn the return from t to t+1.
+    daily = (
+        (weights * fwd).sum(axis=1, min_count=1)
+        / gross.replace(0.0, np.nan)
+    )
     daily = daily.dropna()
     if len(daily) < 2:
+        empty["n_dates"] = int(len(daily))
         return empty
 
     cumulative_return = float((1.0 + daily).prod() - 1.0)