# source: https://raw.githubusercontent.com/remiotore/freqtrade/44beaeb6a420cd8e9f2e4ea93e11d6cfa192ee03/strategies/Discord_MarkovV4.py
# --- Do not remove these libs ---
from freqtrade.strategy.interface import IStrategy
from typing import Dict, List
from pandas import DataFrame
# --------------------------------
import talib.abstract as ta
import freqtrade.vendor.qtpylib.indicators as qtpylib
import numpy as np
import pandas as pd
import statsmodels.api as sm
import os
import time

from pandas.core.common import SettingWithCopyWarning
from statsmodels.tools.sm_exceptions import EstimationWarning
from statsmodels.tools.sm_exceptions import ConvergenceWarning

import warnings
warnings.simplefilter(action="ignore", category=SettingWithCopyWarning)
warnings.simplefilter(action="ignore", category=EstimationWarning)
warnings.simplefilter(action="ignore", category=ConvergenceWarning)

class Github_remiotore_freqtrade__Discord_MarkovV4__20260111_210550(IStrategy):
    stoploss = -0.05

    # Optimal timeframe for the strategy
    timeframe = '30m'

    # run 'populate_indicators' only for new candle
    process_only_new_candles = True

    # Experimental settings (configuration will overide these if set)
    use_sell_signal = True
    sell_profit_only = False
    ignore_roi_if_buy_signal = False

    def bot_loop_start(self, **kwargs):
        headers = ["date", "open", "high", "low", "close", "volume"]
        filename = os.path.join("user_data/data/binance/BTC_USDT-30m.json")
        data = pd.read_json(filename)
        data.columns = headers

        data['returns'] = np.log(data['close'] / data['close'].shift())

        model = sm.tsa.MarkovRegression(data['returns'][-1000:], k_regimes=3, switching_variance=True)
        
        np.random.seed(123)		
        res_1 = model.fit(search_reps=50)

    def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
	
        dataframe['volumeGap'] = np.log(dataframe['volume'] / dataframe['volume'].shift())
        dataframe['dailyChange'] = (dataframe['close'] - dataframe['open']) / dataframe['open']
        dataframe['fractHigh'] = (dataframe['high'] - dataframe['open']) / dataframe['open']
        dataframe['fractLow'] = (dataframe['open'] - dataframe['low']) / dataframe['open']
        dataframe['forecastVariable'] = dataframe['close'].shift(-1) - dataframe['close']

        #dataframe.dropna(inplace=True)
        dataframe = dataframe.fillna(0)
        #dataframe = dataframe[~dataframe.isin([np.nan, np.inf, -np.inf]).any(1)]

        endog = dataframe['forecastVariable']
        exog = dataframe[['volumeGap', 'dailyChange', 'fractHigh', 'fractLow']]

        # Fit the 3-regime model
        mod_2 = sm.tsa.MarkovRegression(endog=endog, k_regimes=3, exog=exog)
        res_2 = mod_2.fit(search_reps=50)

        dataframe['prob_down'] = res_2.smoothed_marginal_probabilities[0]
        dataframe['prob_side'] = res_2.smoothed_marginal_probabilities[1]
        dataframe['prob_up'] = res_2.smoothed_marginal_probabilities[2]

        return dataframe

    def populate_buy_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        dataframe.loc[
            (
                (dataframe['prob_up'] > 0.8)
                &
                (dataframe['prob_side'] < 0.5)
                &
                (dataframe['prob_down'] < 0.5)

            ),
            'buy'] = 1

        return dataframe

    def populate_sell_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        dataframe.loc[
            (
                (dataframe['prob_up'] < 0.5)
                &
                (dataframe['prob_side'] < 0.5)
                &
                (dataframe['prob_down'] > 0.8)

            ),
            'sell'] = 1
        return dataframe