하루y

RSI_SMA: RSI 단순이동평균 지표. 커틀러RSI라고 함.

 - backtrader URL: https://www.backtrader.com/

 - 참조서적: http://www.yes24.com/Product/Goods/90578506

backtrader 지원 지표 (bt.indicators 패키지)

 - Indicator Reference : https://www.backtrader.com/docu/indautoref/

 - 커뮤니티: https://community.backtrader.com/

• Accdecoscillator
• ATR
• Bollinger
• CCI
• Crossover
• Deviation
• DirectionalMove
• DMA
• EMA
• Ichimoku (일목균형표)
• MACD
• Momentum
• Sma: Simple Moving Average
• Stochastic
• Willams
• WMA

아래 샘플에서 차트를 표시하려면, matplotlib버전을 3.2.2 로 내려야 합니다. !!!

pip uninstall matplotlib
pip instaill matplotlib==3.3.2

import backtrader as bt
import yfinance as yf

class MyStrategy(bt.Strategy):
    def __init__(self):
        self.dataclose = self.datas[0].close
        self.order = None
        self.buyprice = None
        self.buycomm = None
        self.rsi = bt.indicators.RSI_SMA(self.data.close, period=21)

    def notify_order(self, order):
        if order.status in [order.Submitted, order.Accepted]:
            return
        if order.status in [order.Completed]:
            if order.isbuy():
                self.log(f'BUY  : 주가 {order.executed.price:,.0f}, '
                    f'수량 {order.executed.size:,.0f}, '
                    f'수수료 {order.executed.comm:,.0f}, '        
                    f'자산 {cerebro.broker.getvalue():,.0f}')
                self.buyprice = order.executed.price
                self.buycomm = order.executed.comm
            else:
                self.log(f'SELL : 주가 {order.executed.price:,.0f}, '
                    f'수량 {order.executed.size:,.0f}, '
                    f'수수료 {order.executed.comm:,.0f}, '
                    f'자산 {cerebro.broker.getvalue():,.0f}')
            self.bar_executed = len(self)
        elif order.status in [order.Canceled]:
            self.log('ORDER CANCELD')
        elif order.status in [order.Margin]:
            self.log('ORDER MARGIN')
        elif order.status in [order.Rejected]:
            self.log('ORDER REJECTED')
        self.order = None

    def next(self):
        if not self.position:
            if self.rsi < 30:
                self.order = self.buy()
        else:
            if self.rsi > 70:
                self.order = self.sell()

    def log(self, txt, dt=None):
        dt = self.datas[0].datetime.date(0)
        print(f'[{dt.isoformat()}] {txt}')

cerebro = bt.Cerebro()
cerebro.addstrategy(MyStrategy)
# data = bt.feeds.YahooFinanceData(dataname='036570.KS', fromdate=datetime(2017, 1, 1), todate=datetime(2019, 12, 1))
data = bt.feeds.PandasData(dataname=yf.download('036570.KS', '2018-01-01', '2021-12-01'))

cerebro.adddata(data)
cerebro.broker.setcash(10000000)
cerebro.broker.setcommission(commission=0.0014)
cerebro.addsizer(bt.sizers.PercentSizer, percents=90)

print(f'Initial Portfolio Value : {cerebro.broker.getvalue():,.0f} KRW')
cerebro.run()
print(f'Final Portfolio Value   : {cerebro.broker.getvalue():,.0f} KRW')
cerebro.plot(style='candlestick')

[*********************100%***********************]  1 of 1 completed
Initial Portfolio Value : 10,000,000 KRW
[2018-03-02] BUY  : 주가 370,000, 수량 24, 수수료 12,432, 자산 10,083,568
[2018-04-02] SELL : 주가 425,000, 수량 -24, 수수료 14,280, 자산 11,293,288
[2018-04-25] BUY  : 주가 358,000, 수량 28, 수수료 14,150, 자산 11,166,205
[2018-09-13] SELL : 주가 427,000, 수량 -28, 수수료 16,878, 자산 13,210,352
[2019-02-22] BUY  : 주가 444,000, 수량 27, 수수료 16,645, 자산 12,832,207
[2019-03-29] SELL : 주가 498,000, 수량 -27, 수수료 18,669, 자산 14,621,035
[2019-06-04] BUY  : 주가 467,500, 수량 28, 수수료 18,522, 자산 14,305,377
[2019-08-05] SELL : 주가 520,000, 수량 -28, 수수료 20,602, 자산 16,067,598
[2019-10-07] BUY  : 주가 509,000, 수량 29, 수수료 20,365, 자산 16,104,390
[2020-01-03] SELL : 주가 547,000, 수량 -29, 수수료 21,886, 자산 17,111,339
[2020-03-20] BUY  : 주가 559,000, 수량 29, 수수료 22,740, 자산 17,321,055
[2020-05-26] SELL : 주가 813,000, 수량 -29, 수수료 33,073, 자산 24,436,002
[2020-08-05] BUY  : 주가 841,000, 수량 26, 수수료 30,607, 자산 24,535,374
[2020-12-14] SELL : 주가 896,000, 수량 -26, 수수료 32,609, 자산 25,802,552
[2021-03-30] BUY  : 주가 834,000, 수량 28, 수수료 32,511, 자산 26,076,330
[2021-11-12] SELL : 주가 757,000, 수량 -28, 수수료 29,510, 자산 23,596,506
Final Portfolio Value   : 23,596,506 KRW

백테스트

특정 투자 전략을 과거 데이터로 검증하는 작업.

백테스트 라이브러리

1. 집라인(Zipline): 대중적으로 많이 사용, 단점 느림. Python버전 환경 영향을 탐. 참조: https://wikidocs.net/4427
   - 재정지원에 중점. 이벤트 기반

   - https://github.com/quantopian/zipline

2. 파이알고트레이드(PyAlgoTrade) (무료) : https://gbeced.github.io/pyalgotrade/

   - 문서: http://gbeced.github.io/pyalgotrade/docs/v0.20/html/

   - 이벤트 기반

3. 트레이딩위드파이썬(TradingWithPython)

4. 파이백테스트(PyBackTest)

5. 백트레이더(BackTrader): 최근 각광, 오픈소스 기반
   https://www.backtrader.com/
 

backtrader 지원 지표 (bt.indicators 패키지)

• Accdecoscillator
• ATR
• Bollinger
• CCI
• Crossover
• Deviation
• DirectionalMove
• DMA
• EMA
• Ichimoku
• MACD
• Momentum
• Sma
• Stochastic
• Willams
• WMA

* 아래 오류 발생 시 matplotlib 버전을 3.2.2 로 재설치 한다.
pip uninstall matplotlib
pip install matplotlib==3.2.2

    from matplotlib.dates import (HOURS_PER_DAY, MIN_PER_HOUR, SEC_PER_MIN,
ImportError: cannot import name 'warnings' from 'matplotlib.dates' (/Users/jeahanjung/opt/anaconda3/envs/system_strading_py39_64/lib/python3.9/site-packages/matplotlib/dates.py)

import backtrader as bt
import yfinance as yf

class RsiStrategy(bt.Strategy):
    def __init__(self):
        self.rsi = bt.indicators.RSI(self.data.close)

    def next(self):
        if not self.position:
            if self.rsi < 30:
                self.order = self.buy()

        elif self.rsi > 70:
            self.order = self.sell()


cerebro = bt.Cerebro()  # create a "Cerebro" engine instance
cerebro.addstrategy(RsiStrategy)  # Add the trading strategy
# data = bt.feeds.YahooFinanceData(dataname="036570.KS", fromdate=datetime(2017, 1, 1), todate=datetime(2019, 12, 1))
data = bt.feeds.PandasData(dataname=yf.download('036570.KS', '2018-01-01', '2021-12-01'))

cerebro.adddata(data)
cerebro.broker.setcash(1000*10000) # 초기투자 금액
cerebro.addsizer(bt.sizers.SizerFix, stake=30) # 매매단위

print(f'Initial Portfolio Value : {cerebro.broker.getvalue():,.0f} KRW')
cerebro.run()  # run it all
print(f'Final Portfolio Value : {cerebro.broker.getvalue():,.0f} KRW')
cerebro.plot()  # and plot it with a sing

[*********************100%***********************]  1 of 1 completed
Initial Portfolio Value : 10,000,000 KRW
Final Portfolio Value : 12,925,000 KRW