Backtest.py

import time
import datetime
import math
import sys
import numpy as np
import pandas as pd
import warnings
warnings.filterwarnings('ignore')
import pyupbit
from Tradingview import Alphatrend as at
from Tradingview import SuperScalper as ss

########################################################################################################################
# Setting values

# You can change market here.
# For example "KRW-XRP" for Ripple, "KRW-SOL" for Solana, "KRW-DOT" for Polkadot.
MARKET = 'KRW-BTC'
# You can change timeframe here.
# For example "minute30" for 30minute, "minute60" for 1hour, "minute240" for 4hour.
TIMEFRAME = 'minute15'
# Candle count in 1 Day
# This value should be reset according to timeframe value you set.
# For example,
# If your timeframe is "minute60", this value should be 24.
# If your timeframe is "minute30", this value should be 48.
# If your timeframe is "minute240", this value should be 6.
DAY_LIMIT = 96

# Enter the number for test days you want.
TEST_DAYS = 30
# Last date of backtest.
# Until this value.
TO_DATE = '2022-08-08 09:00' # 테스트 기간의 마지막 기간

# First starting seed money
SEED_MONEY = 100000 # KRW(Won)

DELAY = 0.5
########################################################################################################################

def Backtest(ticker):
    positionFlag = 0
    orderPrice = 0
    orderSize = 0
    orderAmount = 0
    winCount = 0
    lossCount = 0
    balance = SEED_MONEY
    winrate = 0

    print(f"[1] Start for get OHLC data from Upbit: {datetime.datetime.now()}")

    ohlcCount = DAY_LIMIT * TEST_DAYS
    df_tz = pyupbit.get_ohlcv(ticker, interval=TIMEFRAME, count=ohlcCount, to=TO_DATE)
    df = df_tz.tz_localize(None)
    time.sleep(DELAY)
    del df['value']

    print(f"[1] End for get OHLC data from Upbit: {datetime.datetime.now()}")

    df['orderPrice'] = ''
    df['orderSize'] = ''
    df['orderAmount'] = ''
    df['result'] = ''
    df['percent'] = ''
    df['resultAmount'] = ''
    df['balance'] = ''
    df['winCount'] = ''
    df['lossCount'] = ''
    df['winrate'] = ''

    print(f"[2] Start for get data from AlphaTrend: {datetime.datetime.now()}")

    at_df = at.AlphaTrend(df)
    df['alphaTrend'] = at_df['AlphaTrend']
    df['buySignalk'] = at_df['buySignalk']
    df['sellSignalk'] = at_df['sellSignalk']

    print(f"[2] End for get data from AlphaTrend: {datetime.datetime.now()}")

    print(f"[2] Start for get data from SuperScalper: {datetime.datetime.now()}")

    ss_df = ss.SuperScalper(df)
    #df['RSILong'] = ss_df['RSILong']
    #df['RSIShort'] = ss_df['RSIShort']
    #df['longCondition'] = ss_df['longCondition']
    #df['shortCondition'] = ss_df['shortCondition']
    df['resultSS'] = ss_df['result']
    #df['stopLoss'] = ss_df['stopLoss']
    #df['takeProfit'] = ss_df['takeProfit']

    print(f"[2] End for get data from SuperScalper: {datetime.datetime.now()}")

    tradeNumber = []
    signal = []
    date = []
    price = []
    size = []
    profitAmount = []
    profitPer = []
    totalAmount = []

    print(f"[3] Start for simulation: {datetime.datetime.now()}")

    tradeNumberCount = 1

    for i in range(1, len(df)):
        #if df['buySignalk'].iloc[i - 1] == 1: # For AlphaTrend
        if df['resultSS'].iloc[i - 1] == 1: # For SuperScalper
            if positionFlag == 0:
                if balance > 0:
                    tradeNumber.append(tradeNumberCount)
                    signal.append('BUY')
                    date.append(df.index[i])
                    df['orderPrice'].iloc[i] = orderPrice = df['open'].iloc[i]
                    price.append(format(orderPrice, ','))
                    df['orderSize'].iloc[i] = orderSize = round(balance / orderPrice, 4)
                    size.append(orderSize)
                    orderAmount = int(balance)
                    df['orderAmount'].iloc[i] = format(orderAmount, ',')
                    profitAmount.append('')
                    profitPer.append('')
                    totalAmount.append('')
                    positionFlag = 1

        #elif df['sellSignalk'].iloc[i - 1] == 1: # For AlphaTrend
        elif df['resultSS'].iloc[i - 1] == 2:  # For SuperScalper
            if positionFlag == 1:
                tradeNumber.append(tradeNumberCount)
                signal.append('SELL')
                date.append(df.index[i])

                if df['open'].iloc[i] > orderPrice:
                    df['result'].iloc[i] = "WIN"
                    winCount += 1
                else:
                    df['result'].iloc[i] = "LOSE"
                    lossCount += 1

                df['winCount'].iloc[i] = winCount
                df['lossCount'].iloc[i] = lossCount
                if winCount > 0:
                    winrate = (winCount / (winCount + lossCount)) * 100
                else:
                    winrate = 0
                winrate = round(winrate, 2)
                df['winrate'].iloc[i] = f"{winrate}%"

                price.append(format(df['open'].iloc[i], ','))
                size.append(orderSize)
                percent = round((abs(df['open'].iloc[i] - orderPrice) / orderPrice) * 100, 2)
                resultAmount = int((balance * (percent / 100)))

                if df['open'].iloc[i] > orderPrice:
                    balance = balance + resultAmount
                    df['percent'].iloc[i] = f"{percent}%"
                    df['resultAmount'].iloc[i] = format(resultAmount, ',')

                    profitAmount.append(format(resultAmount, ','))
                    profitPer.append(f'{percent}%')
                else:
                    balance = balance - resultAmount
                    df['percent'].iloc[i] = f"-{percent}%"
                    df['resultAmount'].iloc[i] = format(-resultAmount, ',')

                    profitAmount.append(format(-resultAmount, ','))
                    profitPer.append(f'-{percent}%')

                balance = int(balance)
                df['balance'].iloc[i] = format(balance, ',')
                totalAmount.append(format(balance, ','))
                positionFlag = 0

                tradeNumberCount += 1

    print(f"[3] End for simulation: {datetime.datetime.now()}")

    print(f"[4] Start for make excel file for result: {datetime.datetime.now()}")

    output_df = pd.DataFrame()
    output_df['#'] = tradeNumber
    output_df['Signal'] = signal
    output_df['Date'] = date
    output_df['Order Price (KRW)'] = price
    output_df['Order Size'] = size
    output_df['Profit or Loss (KRW)'] = profitAmount
    output_df['Profit or Loss (%)'] = profitPer
    output_df['Balance'] = totalAmount

    title = ticker
    output_df.to_excel(f'Output/{title}_Backtest_Result.xlsx')

    print(f"[4] End for make excel file for result: {datetime.datetime.now()}")

########################################################################################################################
print(f"Backtest: {MARKET}")
Backtest(ticker=MARKET)