Initial commit: cc-cursor 全链路量化研究平台

7 Sprints 全部完成: Sprint 0: 基础设施 (DataManager + MariaDB) Sprint 1: 因子引擎 (34因子/12分类) Sprint 2: VectorBT 回测 (5策略+截面) Sprint 3: Optuna 优化 (+Walk-Forward) Sprint 4: ML 模型 (LightGBM+CatBoost) Sprint 5: Qwen 情绪因子 (三源新闻+日期对齐) Sprint 6: Agent 系统 (4Agent+日报.md/.html) 生产加固 (15项): Tushare双源fallback, SSH自动恢复, pool_pre_ping, save_daily先删后插, load_dotenv绝对路径, 日报5d/20d修复, RiskAgent改上证指数, 昨日对比+数据截止, mac_report utf8mb4, CLAUDE-*.md 9条已知Bug, demo全参数化, djapi数据源归一化, indexDatas API修正 Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-07 15:59:05 +08:00
commit 271a9343a5
293 changed files with 59598 additions and 0 deletions
@@ -0,0 +1,24 @@
+"""
+振幅因子。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class AmplitudeFactor(BaseFactor):
+    """N 日均振幅 = mean((high - low) / close, N) * 100"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 5):
+        self.period = period
+        self.name = f"amplitude_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        daily_amp = (df["high"] - df["low"]) / df["close"].replace(0, float("nan")) * 100
+        return daily_amp.rolling(window=self.period, min_periods=self.period).mean()
+
+    def get_required_columns(self) -> list[str]:
+        return ["high", "low", "close"]
@@ -0,0 +1,47 @@
+"""
+ATR 平均真实波幅因子。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class ATRFactor(BaseFactor):
+    """Average True Range，衡量波动性。"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 14):
+        self.period = period
+        self.name = f"atr_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        high, low, close = df["high"], df["low"], df["close"]
+        prev_close = close.shift(1)
+        tr = pd.concat([
+            (high - low).abs(),
+            (high - prev_close).abs(),
+            (low - prev_close).abs(),
+        ], axis=1).max(axis=1)
+        return tr.ewm(span=self.period, min_periods=self.period).mean()
+
+    def get_required_columns(self) -> list[str]:
+        return ["high", "low", "close"]
+
+
+class ATRRatioFactor(BaseFactor):
+    """ATR / close 归一化，便于跨股票比较。"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 14):
+        self.period = period
+        self.name = f"atr_ratio_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        atr = ATRFactor(period=self.period).calculate(df)
+        return atr / df["close"].replace(0, float("nan")) * 100
+
+    def get_required_columns(self) -> list[str]:
+        return ["high", "low", "close"]
@@ -0,0 +1,53 @@
+"""
+布林带因子：价格在布林带中的位置。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class BollingerFactor(BaseFactor):
+    """
+    布林带位置 = (close - middle) / (upper - lower)
+
+    值在 0~1 之间：接近 0 表示在下轨，接近 1 表示在上轨。
+    """
+
+    category = "technical"
+
+    def __init__(self, period: int = 20, std: float = 2.0):
+        self.period = period
+        self.std = std
+        self.name = f"boll_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        middle = df["close"].rolling(window=self.period, min_periods=self.period).mean()
+        std = df["close"].rolling(window=self.period, min_periods=self.period).std()
+        upper = middle + self.std * std
+        lower = middle - self.std * std
+        band_width = upper - lower
+        return ((df["close"] - lower) / band_width.replace(0, float("nan"))).clip(0, 1)
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
+
+
+class BollingerWidthFactor(BaseFactor):
+    """布林带宽度 = (upper - lower) / middle * 100"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 20, std: float = 2.0):
+        self.period = period
+        self.std = std
+        self.name = f"boll_width_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        middle = df["close"].rolling(window=self.period, min_periods=self.period).mean()
+        std = df["close"].rolling(window=self.period, min_periods=self.period).std()
+        band_width = 2 * self.std * std
+        return band_width / middle.replace(0, float("nan")) * 100
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
@@ -0,0 +1,49 @@
+"""
+均线交叉因子。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class MACrossFactor(BaseFactor):
+    """
+    均线交叉信号。
+
+    返回：fast_ma / slow_ma - 1，正值表示短期均线在上方。
+    """
+
+    category = "technical"
+
+    def __init__(self, fast: int = 5, slow: int = 20):
+        self.fast = fast
+        self.slow = slow
+        self.name = f"ma_cross_{fast}_{slow}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        ma_fast = df["close"].rolling(window=self.fast, min_periods=self.fast).mean()
+        ma_slow = df["close"].rolling(window=self.slow, min_periods=self.slow).mean()
+        return (ma_fast / ma_slow.replace(0, float("nan"))) - 1
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
+
+
+class MADeviationFactor(BaseFactor):
+    """
+    价格偏离均线程度 = (close - ma) / ma * 100
+    """
+
+    category = "technical"
+
+    def __init__(self, period: int = 20):
+        self.period = period
+        self.name = f"ma_dev_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        ma = df["close"].rolling(window=self.period, min_periods=self.period).mean()
+        return (df["close"] - ma) / ma.replace(0, float("nan")) * 100
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
@@ -0,0 +1,47 @@
+"""
+MACD 因子。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class MACDFactor(BaseFactor):
+    """
+    MACD 系列因子。
+
+    返回 DIF/DEA/HIST 三个值。使用 calculate() 返回 HIST（柱），
+    单独方法获取 DIF/DEA。
+    """
+
+    category = "technical"
+
+    def __init__(self, fast: int = 12, slow: int = 26, signal: int = 9):
+        self.fast = fast
+        self.slow = slow
+        self.signal = signal
+        self.name = f"macd_{fast}_{slow}_{signal}"
+
+    def _ema(self, series: pd.Series, span: int) -> pd.Series:
+        return series.ewm(span=span, min_periods=span).mean()
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        """返回 MACD 柱（DIF - DEA）。"""
+        ema_fast = self._ema(df["close"], self.fast)
+        ema_slow = self._ema(df["close"], self.slow)
+        dif = ema_fast - ema_slow
+        dea = self._ema(dif, self.signal)
+        return dif - dea
+
+    def dif(self, df: pd.DataFrame) -> pd.Series:
+        ema_fast = self._ema(df["close"], self.fast)
+        ema_slow = self._ema(df["close"], self.slow)
+        return ema_fast - ema_slow
+
+    def dea(self, df: pd.DataFrame) -> pd.Series:
+        dif = self.dif(df)
+        return self._ema(dif, self.signal)
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
@@ -0,0 +1,23 @@
+"""
+动量因子：N 日收益率。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class MomentumFactor(BaseFactor):
+    """N 日价格动量 = (close_t - close_{t-N}) / close_{t-N} * 100"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 20):
+        self.period = period
+        self.name = f"momentum_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        return df["close"].pct_change(periods=self.period) * 100
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
@@ -0,0 +1,29 @@
+"""
+RSI 相对强弱因子。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class RSIFactor(BaseFactor):
+    """Wilder's RSI = 100 - 100 / (1 + RS), RS = avg_gain / avg_loss"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 14):
+        self.period = period
+        self.name = f"rsi_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        delta = df["close"].diff()
+        gain = delta.clip(lower=0)
+        loss = (-delta).clip(lower=0)
+        avg_gain = gain.ewm(span=self.period, min_periods=self.period).mean()
+        avg_loss = loss.ewm(span=self.period, min_periods=self.period).mean()
+        rs = avg_gain / avg_loss.replace(0, float("nan"))
+        return 100 - 100 / (1 + rs)
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
@@ -0,0 +1,40 @@
+"""
+换手率因子。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class TurnoverFactor(BaseFactor):
+    """N 日均换手率。"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 5):
+        self.period = period
+        self.name = f"turnover_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        return df["turnover_rate"].rolling(window=self.period, min_periods=self.period).mean()
+
+    def get_required_columns(self) -> list[str]:
+        return ["turnover_rate"]
+
+
+class TurnoverChangeFactor(BaseFactor):
+    """换手率变化 = 当日换手率 / N 日均换手率。"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 5):
+        self.period = period
+        self.name = f"turnover_chg_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        avg = df["turnover_rate"].rolling(window=self.period, min_periods=self.period).mean()
+        return df["turnover_rate"] / avg.replace(0, float("nan"))
+
+    def get_required_columns(self) -> list[str]:
+        return ["turnover_rate"]
@@ -0,0 +1,42 @@
+"""
+波动率因子。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class VolatilityFactor(BaseFactor):
+    """N 日年化波动率 = std(daily_return, N) * sqrt(252) * 100"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 20):
+        self.period = period
+        self.name = f"volatility_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        daily_ret = df["close"].pct_change()
+        return daily_ret.rolling(window=self.period, min_periods=self.period).std() * (252 ** 0.5) * 100
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
+
+
+class DownsideVolatilityFactor(BaseFactor):
+    """下行波动率：只计算负收益的标准差。"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 20):
+        self.period = period
+        self.name = f"down_vol_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        daily_ret = df["close"].pct_change()
+        downside = daily_ret.clip(upper=0)
+        return downside.rolling(window=self.period, min_periods=self.period).std() * (252 ** 0.5) * 100
+
+    def get_required_columns(self) -> list[str]:
+        return ["close"]
@@ -0,0 +1,40 @@
+"""
+量价因子：量比 / 成交量变化率。
+"""
+
+import pandas as pd
+
+from factors.base import BaseFactor
+
+
+class VolumeFactor(BaseFactor):
+    """N 日均量比 = vol / mean(vol, N)"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 5):
+        self.period = period
+        self.name = f"vol_ratio_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        avg_vol = df["vol"].rolling(window=self.period, min_periods=self.period).mean()
+        return df["vol"] / avg_vol.replace(0, float("nan"))
+
+    def get_required_columns(self) -> list[str]:
+        return ["vol"]
+
+
+class VolumeChangeFactor(BaseFactor):
+    """成交量 N 日变化率"""
+
+    category = "technical"
+
+    def __init__(self, period: int = 5):
+        self.period = period
+        self.name = f"vol_chg_{period}"
+
+    def calculate(self, df: pd.DataFrame) -> pd.Series:
+        return df["vol"].pct_change(periods=self.period) * 100
+
+    def get_required_columns(self) -> list[str]:
+        return ["vol"]