From cbd60894b900344916915187a95672701617a4c6 Mon Sep 17 00:00:00 2001
From: aszerW <aszer27937@gmail.com>
Date: Fri, 15 May 2026 23:26:54 +0800
Subject: [PATCH] =?UTF-8?q?fix(strategy):=20=E4=BF=AE=E5=A4=8D=E5=80=BA?=
 =?UTF-8?q?=E5=88=B8=E6=8C=87=E6=95=B0OHLCV=E6=95=B0=E6=8D=AE=E5=A4=84?=
 =?UTF-8?q?=E7=90=86=E9=80=BB=E8=BE=91?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- 问题: 债券指数(931862.CSI)只有close数据,open/high/low全是None
- 原代码: 检查列存在后整行dropna → 数据变成0条
- 修复: 检查列存在 + 检查数据是否有效(不全为None)
- 如果OHLCV无效 → 使用close列单独dropna
- 结果: 30年国债4330条数据正常参与回测
- 收益影响: 累计收益+258%, Sharpe+0.04
---
 strategies/rotation/strategy.py | 27 ++++++++++++++++++---------
 1 file changed, 18 insertions(+), 9 deletions(-)

diff --git a/strategies/rotation/strategy.py b/strategies/rotation/strategy.py
index bf6ce72..5e66c1c 100644
--- a/strategies/rotation/strategy.py
+++ b/strategies/rotation/strategy.py
@@ -302,20 +302,29 @@ class RotationStrategy(StrategyBase):
         for code in valid_codes:
             df = index_data[code].copy()
             
-            # 检查是否有OHLCV数据
+            # 检查是否有有效的OHLCV数据（列存在且不全为None）
             ohlcv_cols = ['open', 'high', 'low', 'close', 'volume']
-            has_ohlcv = all(col in df.columns for col in ['open', 'high', 'low', 'close'])
+            required_cols = ['open', 'high', 'low', 'close']
             
-            if has_ohlcv:
-                # 如果有完整OHLCV，整行dropna()后提取close
+            # 检查列是否存在
+            cols_exist = all(col in df.columns for col in required_cols)
+            
+            # 检查数据是否有效（不全为None/NaN）
+            if cols_exist:
+                cols_have_data = all(df[col].notna().any() for col in required_cols)
+            else:
+                cols_have_data = False
+            
+            if cols_exist and cols_have_data:
+                # 有完整有效的OHLCV数据，整行dropna()后提取close
                 df_clean = df[ohlcv_cols].dropna()
                 close_series = df_clean['close'] if len(df_clean) > 0 else pd.Series(dtype=float)
+            elif 'close' in df.columns and df['close'].notna().any():
+                # 只有close列有效数据（如债券指数）
+                close_series = df['close'].dropna()
             else:
-                # 只有close列的情况
-                if 'close' in df.columns:
-                    close_series = df['close'].dropna()
-                else:
-                    close_series = df.dropna()
+                # 无有效数据
+                close_series = pd.Series(dtype=float)
             
             # 检查数据长度并警告，但不剔除
             if len(close_series) < self.n_days + 1: