Strategy07_TEMA

# pragma pylint: disable=missing-docstring, invalid-name, pointless-string-statement
# flake8: noqa: F401

"""
Strategy07_TEMA
===============
A Triple Exponential Moving Average (TEMA) crossover strategy with
Stochastic Oscillator confirmation, designed for SPOT crypto trading.
Halal-compliant: SPOT only — no shorting, no margin, no futures.

Concept
-------
TEMA eliminates the lag inherent in simple and double EMAs by applying
a triple-smoothing formula:

  TEMA = 3 × EMA1 - 3 × EMA2 + EMA3

where EMA1 = EMA(close), EMA2 = EMA(EMA1), EMA3 = EMA(EMA2).

TEMA responds to price changes faster than EMA or DEMA while keeping
noise lower than a raw price signal. The crossover of a fast TEMA (10)
over a slow TEMA (50) identifies trend direction changes early.

Entry Logic (3-step confirmation)
-----------------------------------
Step 1 — Crossover detection:
  Fast TEMA (10) crosses ABOVE slow TEMA (50).
  This is the initial trend-flip signal.

Step 2 — Bullish candle confirmation (2 bars):
  The 2 candles immediately following the crossover must BOTH close
  bullish (close > open). This filters out whipsaws — if the crossover
  was genuine, price should continue higher with bullish momentum.

Step 3 — Volume and Stochastic filters:
  • Volume > 20-period average      → confirms real buying interest.
  • Stochastic %K (14,3,3) < 20     → price is oversold at crossover,
    providing a better risk/reward entry point.

Exit Logic (2-bar confirmation)
---------------------------------
Fast TEMA crosses BELOW slow TEMA AND the next 2 candles are BOTH
bearish (close < open). Mirror of entry — avoids exiting on brief
pullbacks within a still-valid uptrend.

Risk Parameters (from hyperopt)
---------------------------------
stoploss           : -2.2 %  (tight — TEMA crossover strategy trades
                               frequently; small losses cut quickly)
trailing_stop      : True
ROI table          : 4 % (0 min) → 3 % (30 min) → 2 % (60 min) → 1 % (120 min)

Timeframe : 4h
Author    : Halal Algo Batch 2
"""

from freqtrade.strategy import IStrategy
import numpy as np
import pandas as pd
import pandas_ta as ta
from pandas import DataFrame


# ---------------------------------------------------------------------------
# Utility helpers
# ---------------------------------------------------------------------------

def crossed_above(s1: pd.Series, s2: pd.Series) -> pd.Series:
    """Return True on the bar where s1 crosses above s2."""
    return (s1 > s2) & (s1.shift(1) <= s2.shift(1))


def crossed_below(s1: pd.Series, s2: pd.Series) -> pd.Series:
    """Return True on the bar where s1 crosses below s2."""
    return (s1 < s2) & (s1.shift(1) >= s2.shift(1))


# ---------------------------------------------------------------------------
# Strategy
# ---------------------------------------------------------------------------

class Strategy07_TEMA(IStrategy):
    """
    TEMA (10/50) Crossover + Stochastic Oversold + Volume Filter
    SPOT only — halal algo trading, no leverage.
    """

    # ------------------------------------------------------------------
    # Freqtrade metadata
    # ------------------------------------------------------------------
    INTERFACE_VERSION = 3
    timeframe = "4h"
    can_short = False  # SPOT only

    # TEMA(50) needs ~150 candles to converge; add buffer for Stochastic(14).
    startup_candle_count: int = 200

    # ------------------------------------------------------------------
    # ROI
    # ------------------------------------------------------------------
    minimal_roi = {
        "0":   0.04,   # 4 % immediately
        "30":  0.03,   # 3 % after 30 min (0.5 x 4h bar)
        "60":  0.02,   # 2 % after 60 min
        "120": 0.01,   # 1 % after 120 min
    }

    # ------------------------------------------------------------------
    # Stop-loss & trailing stop
    # ------------------------------------------------------------------
    stoploss = -0.022              # -2.2 % tight stop (hyperopt result)
    trailing_stop = True
    trailing_stop_positive = 0.01  # activate trailing stop once +1 % profit
    trailing_stop_positive_offset = 0.02
    trailing_only_offset_is_reached = True

    # ------------------------------------------------------------------
    # Misc
    # ------------------------------------------------------------------
    process_only_new_candles = True
    use_exit_signal = True
    exit_profit_only = False
    ignore_roi_if_entry_signal = False

    # ------------------------------------------------------------------
    # Indicator calculation
    # ------------------------------------------------------------------

    def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        """
        Compute TEMA fast/slow, Stochastic oscillator, and volume MA.

        TEMA (Triple Exponential Moving Average)
        -----------------------------------------
        pandas_ta computes TEMA using the exact triple-EMA formula.
        Fast (10) reacts sharply to recent price moves.
        Slow (50) provides a stable trend baseline.

        Stochastic Oscillator (14, 3, 3)
        ----------------------------------
        %K : raw stochastic — position of close within the 14-bar range.
        %D : 3-period SMA of %K (signal line).
        Values < 20 indicate oversold; values > 80 indicate overbought.
        We use %K < 20 as an entry filter for better risk/reward entries.

        Crossover signals
        ------------------
        tema_cross_up   : fast TEMA crosses above slow TEMA (potential entry)
        tema_cross_down : fast TEMA crosses below slow TEMA (potential exit)

        2-bar confirmation columns
        ---------------------------
        bull_confirm  : 2 consecutive bullish candles (close > open) after crossover
        bear_confirm  : 2 consecutive bearish candles (close < open) after crossover
        """

        # ── TEMA indicators ─────────────────────────────────────────────
        dataframe["tema_fast"] = ta.tema(dataframe["close"], length=10)
        dataframe["tema_slow"] = ta.tema(dataframe["close"], length=50)

        # ── Crossover events ────────────────────────────────────────────
        dataframe["tema_cross_up"]   = crossed_above(dataframe["tema_fast"], dataframe["tema_slow"])
        dataframe["tema_cross_down"] = crossed_below(dataframe["tema_fast"], dataframe["tema_slow"])

        # ── 2-bar bullish confirmation after crossover ──────────────────
        # A "bullish candle" is one where close > open.
        candle_bull = (dataframe["close"] > dataframe["open"]).astype(int)

        # We want: crossover happened 2 bars ago, AND candles at t-1 AND t are bull.
        # This checks that the 2 bars immediately AFTER the cross are both bullish.
        # At the time of signal generation (bar t):
        #   - bar t-2 is when the cross happened
        #   - bar t-1 must be bullish
        #   - bar t   must be bullish
        dataframe["bull_confirm"] = (
            dataframe["tema_cross_up"].shift(2).fillna(False).astype(bool) &
            (candle_bull.shift(1) == 1) &
            (candle_bull == 1)
        )

        # ── 2-bar bearish confirmation after crossover ──────────────────
        candle_bear = (dataframe["close"] < dataframe["open"]).astype(int)
        dataframe["bear_confirm"] = (
            dataframe["tema_cross_down"].shift(2).fillna(False).astype(bool) &
            (candle_bear.shift(1) == 1) &
            (candle_bear == 1)
        )

        # ── Stochastic Oscillator (14, 3, 3) ────────────────────────────
        stoch = ta.stoch(
            dataframe["high"], dataframe["low"], dataframe["close"],
            k=14, d=3, smooth_k=3
        )
        # pandas_ta column names: STOCHk_14_3_3, STOCHd_14_3_3
        dataframe["slowk"] = stoch["STOCHk_14_3_3"]
        dataframe["slowd"] = stoch["STOCHd_14_3_3"]

        # ── Volume filter : 20-period SMA of volume ─────────────────────
        dataframe["vol_ma20"] = ta.sma(dataframe["volume"], length=20)

        # --- NaN Safety: convert any None to NaN so pandas comparisons work ---
        for col in dataframe.columns:
            if col not in ['open', 'high', 'low', 'close', 'volume', 'date']:
                try:
                    dataframe[col] = pd.to_numeric(dataframe[col], errors='coerce')
                except (ValueError, TypeError):
                    pass

        return dataframe

    # ------------------------------------------------------------------
    # Entry signal
    # ------------------------------------------------------------------

    def populate_entry_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        """
        Long entry conditions (ALL must be true simultaneously):

        1. bull_confirm = True
           → TEMA fast crossed above TEMA slow exactly 2 bars ago, AND
             both the bar after the cross AND the current bar are bullish
             (close > open). Two confirmed bullish candles validate the
             crossover is not a whipsaw.

        2. Volume > 20-period average
           → Confirms real buying interest is behind the move.

        3. Stochastic %K < 20 (oversold)
           → Entry price is at a relatively low point within recent range.
             This improves risk/reward and reduces the chance of buying
             at a local top right after the TEMA crossover.

        4. Volume > 0 (data quality guard).
        """
        conditions = (
            dataframe["bull_confirm"] &
            (dataframe["volume"] > dataframe["vol_ma20"]) &
            (dataframe["slowk"] < 20) &
            (dataframe["volume"] > 0)
        )

        dataframe.loc[conditions, "enter_long"] = 1
        return dataframe

    # ------------------------------------------------------------------
    # Exit signal
    # ------------------------------------------------------------------

    def populate_exit_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        """
        Long exit conditions (ALL must be true simultaneously):

        1. bear_confirm = True
           → TEMA fast crossed below TEMA slow exactly 2 bars ago, AND
             both the bar after the cross AND the current bar are bearish
             (close < open). Two confirmed bearish candles validate the
             crossover is a genuine trend reversal, not a brief pullback.

        2. Volume > 0 (data quality guard).

        Note: The tight -2.2 % stoploss and trailing stop provide additional
        protection for sharp drops that do not wait for a clean TEMA crossover.
        """
        conditions = (
            dataframe["bear_confirm"] &
            (dataframe["volume"] > 0)
        )

        dataframe.loc[conditions, "exit_long"] = 1
        return dataframe