Create indicators.py

indicators.py

@@ -0,0 +1,215 @@
+# indicators.py
+import pandas as pd
+import numpy as np
+
+# Try TA-Lib
+try:
+    import talib
+    TALIB = True
+except:
+    TALIB = False
+
+
+# ============================================================
+# BASIC INDICATORS (SMA, EMA, ATR)
+# ============================================================
+
+def sma(series, period):
+    return series.rolling(period).mean()
+
+
+def ema(series, period):
+    return series.ewm(span=period, adjust=False).mean()
+
+
+def atr(high, low, close, period=14):
+    if TALIB and hasattr(talib, "ATR"):
+        return talib.ATR(high, low, close, timeperiod=period)
+    else:
+        tr1 = high - low
+        tr2 = (high - close.shift()).abs()
+        tr3 = (low - close.shift()).abs()
+        tr = pd.concat([tr1, tr2, tr3], axis=1).max(axis=1)
+        return tr.rolling(period).mean()
+
+
+# ============================================================
+# SUPERTREND — TradingView Perfect Replication
+# ============================================================
+
+def supertrend(df, period=10, multiplier=3):
+    """
+    Returns:
+    ST: SuperTrend line
+    DIR: Trend direction (True = Uptrend, False = Downtrend)
+    """
+    high = df['High']
+    low = df['Low']
+    close = df['Close']
+
+    # ATR
+    atr_val = atr(high, low, close, period)
+
+    # Basic bands
+    hl2 = (high + low) / 2
+    upperband = hl2 + multiplier * atr_val
+    lowerband = hl2 - multiplier * atr_val
+
+    final_upper = upperband.copy()
+    final_lower = lowerband.copy()
+
+    for i in range(1, len(df)):
+        # Final upper band
+        if upperband.iloc[i] < final_upper.iloc[i - 1] or close.iloc[i - 1] > final_upper.iloc[i - 1]:
+            final_upper.iloc[i] = upperband.iloc[i]
+        else:
+            final_upper.iloc[i] = final_upper.iloc[i - 1]
+
+        # Final lower band
+        if lowerband.iloc[i] > final_lower.iloc[i - 1] or close.iloc[i - 1] < final_lower.iloc[i - 1]:
+            final_lower.iloc[i] = lowerband.iloc[i]
+        else:
+            final_lower.iloc[i] = final_lower.iloc[i - 1]
+
+    # Supertrend
+    st = pd.Series(index=df.index)
+    dir_up = pd.Series(index=df.index, dtype=bool)
+
+    for i in range(1, len(df)):
+        if close.iloc[i] > final_upper.iloc[i - 1]:
+            dir_up.iloc[i] = True
+        elif close.iloc[i] < final_lower.iloc[i - 1]:
+            dir_up.iloc[i] = False
+        else:
+            dir_up.iloc[i] = dir_up.iloc[i - 1]
+
+        st.iloc[i] = final_lower.iloc[i] if dir_up.iloc[i] else final_upper.iloc[i]
+
+    return st, dir_up
+
+
+# ============================================================
+# KELTNER CHANNEL — (EMA ± ATR * Mult)
+# ============================================================
+
+def keltner_channel(df, ema_period=20, atr_period=10, mult=2):
+    close = df['Close']
+    upper = ema(close, ema_period) + mult * atr(df['High'], df['Low'], df['Close'], atr_period)
+    lower = ema(close, ema_period) - mult * atr(df['High'], df['Low'], df['Close'], atr_period)
+    mid = ema(close, ema_period)
+    return mid, upper, lower
+
+
+# ============================================================
+# ZIGZAG — PERCENT BASED
+# ============================================================
+
+def zigzag(series, pct=5):
+    """
+    Returns ZigZag turning points based on percentage move.
+    """
+    zz = pd.Series(index=series.index, dtype=float)
+    last_pivot_price = series.iloc[0]
+    last_pivot_idx = series.index[0]
+    trend = 0  # +1 up, -1 down
+
+    for i in range(1, len(series)):
+        change = (series.iloc[i] - last_pivot_price) / last_pivot_price * 100
+
+        if trend >= 0 and change <= -pct:
+            zz.iloc[i] = series.iloc[i]
+            last_pivot_price = series.iloc[i]
+            trend = -1
+
+        elif trend <= 0 and change >= pct:
+            zz.iloc[i] = series.iloc[i]
+            last_pivot_price = series.iloc[i]
+            trend = +1
+
+    return zz
+
+
+# ============================================================
+# SWING HIGH / SWING LOW
+# ============================================================
+
+def swing_high_low(df, window=5):
+    """
+    Identifies swing high/low using window method.
+    For window=5, center index is 2 (2 left, 2 right)
+    """
+    highs = df['High']
+    lows = df['Low']
+    idx = df.index
+
+    swing_high = pd.Series(np.nan, index=idx)
+    swing_low = pd.Series(np.nan, index=idx)
+
+    half = window // 2
+
+    for i in range(half, len(df) - half):
+        segment_high = highs[i - half: i + half + 1]
+        segment_low = lows[i - half: i + half + 1]
+
+        if highs.iloc[i] == segment_high.max():
+            swing_high.iloc[i] = highs.iloc[i]
+
+        if lows.iloc[i] == segment_low.min():
+            swing_low.iloc[i] = lows.iloc[i]
+
+    return swing_high, swing_low
+
+
+# ============================================================
+# RSI FALLBACK
+# ============================================================
+
+def rsi(series, period=14):
+    if TALIB and hasattr(talib, "RSI"):
+        return talib.RSI(series, timeperiod=period)
+    else:
+        delta = series.diff()
+        up = delta.clip(lower=0)
+        down = -delta.clip(upper=0)
+        ema_up = up.ewm(span=period).mean()
+        ema_down = down.ewm(span=period).mean()
+        rs = ema_up / ema_down
+        return 100 - (100 / (1 + rs))
+
+
+# ============================================================
+# MACD FALLBACK
+# ============================================================
+
+def macd(series, fast=12, slow=26, signal=9):
+    if TALIB and hasattr(talib, "MACD"):
+        macd_line, macd_signal, macd_hist = talib.MACD(series, fastperiod=fast, slowperiod=slow, signalperiod=signal)
+        return macd_line, macd_signal, macd_hist
+    else:
+        ema_fast = ema(series, fast)
+        ema_slow = ema(series, slow)
+        macd_line = ema_fast - ema_slow
+        macd_signal = ema(macd_line, signal)
+        macd_hist = macd_line - macd_signal
+        return macd_line, macd_signal, macd_hist
+
+
+# ============================================================
+# STOCHASTIC FALLBACK
+# ============================================================
+
+def stochastic(df, k_period=14, d_period=3):
+    if TALIB and hasattr(talib, "STOCH"):
+        k, d = talib.STOCH(
+            df['High'], df['Low'], df['Close'],
+            fastk_period=k_period,
+            slowk_period=d_period, slowk_matype=0,
+            slowd_period=d_period, slowd_matype=0
+        )
+        return k, d
+    else:
+        low_min = df['Low'].rolling(k_period).min()
+        high_max = df['High'].rolling(k_period).max()
+        k = (df['Close'] - low_min) * 100 / (high_max - low_min)
+        d = k.rolling(d_period).mean()
+        return k, d