shortcash.ipynb

Kernel: Python 3

from datetime import datetime
import backtrader as bt
import pandas as pd
import numpy as np
import vectorbt as vbt

df = pd.DataFrame(index=[datetime(2020, 1, i + 1) for i in range(9)])
df['open'] = [1, 1, 2, 3, 4, 5, 6, 7, 8]
df['high'] = df['open'] + 0.5
df['low'] = df['open'] - 0.5
df['close'] = df['open']
data = bt.feeds.PandasData(dataname=df)
size = np.array([5, 5, -5, -5, -5, -5, 5, 5, 0])


class CommInfoFloat(bt.CommInfoBase):
    """Commission schema that keeps size as float."""
    params = (
        ('stocklike', True),
        ('commtype', bt.CommInfoBase.COMM_PERC),
        ('percabs', True),
      )
    
    def getsize(self, price, cash):
        if not self._stocklike:
            return self.p.leverage * (cash / self.get_margin(price))

        return self.p.leverage * (cash / price)


class CashValueAnalyzer(bt.analyzers.Analyzer):
    """Analyzer to extract cash and value."""
    def create_analysis(self):
        self.rets = {}

    def notify_cashvalue(self, cash, value):
        self.rets[self.strategy.datetime.datetime()] = (cash, value)

    def get_analysis(self):
        return self.rets


class TestStrategy(bt.Strategy):
    def __init__(self):
        self.i = 0
        
    def log(self, txt, dt=None):
        dt = dt or self.data.datetime[0]
        dt = bt.num2date(dt)
        print('%s, %s' % (dt.isoformat(), txt))
        
    def notify_order(self, order):
        if order.status in [bt.Order.Submitted, bt.Order.Accepted]:
            return  # Await further notifications

        if order.status == order.Completed:
            if order.isbuy():
                buytxt = 'BUY COMPLETE {}, size = {:.2f}, price = {:.2f}'.format(
                    order.data._name, order.executed.size, order.executed.price)
                self.log(buytxt, order.executed.dt)
            else:
                selltxt = 'SELL COMPLETE {}, size = {:.2f}, price = {:.2f}'.format(
                    order.data._name, order.executed.size, order.executed.price)
                self.log(selltxt, order.executed.dt)

        elif order.status in [order.Expired, order.Canceled, order.Margin]:
            self.log('%s ,' % order.Status[order.status])
            pass  # Simply log

        # Allow new orders
        self.orderid = None
        
    def next(self):
        if size[self.i] > 0:
            self.buy(size=size[self.i])
        elif size[self.i] < 0:
            self.sell(size=-size[self.i])
        self.i += 1

def bt_simulate(shortcash):
    cerebro = bt.Cerebro()
    comminfo = CommInfoFloat(commission=0.01)
    cerebro.broker.addcommissioninfo(comminfo)
    cerebro.addstrategy(TestStrategy)
    cerebro.addanalyzer(CashValueAnalyzer)
    cerebro.broker.setcash(100.)
    cerebro.broker.set_checksubmit(False)
    cerebro.broker.set_shortcash(shortcash)
    cerebro.adddata(data)
    return cerebro.run()[0]

strategy = bt_simulate(True)
strategy.analyzers.cashvalueanalyzer.get_analysis()

2020-01-02T00:00:00, BUY COMPLETE , size = 5.00, price = 1.00
2020-01-03T00:00:00, BUY COMPLETE , size = 5.00, price = 2.00
2020-01-04T00:00:00, SELL COMPLETE , size = -5.00, price = 3.00
2020-01-05T00:00:00, SELL COMPLETE , size = -5.00, price = 4.00
2020-01-06T00:00:00, SELL COMPLETE , size = -5.00, price = 5.00
2020-01-07T00:00:00, SELL COMPLETE , size = -5.00, price = 6.00
2020-01-08T00:00:00, BUY COMPLETE , size = 5.00, price = 7.00
2020-01-09T00:00:00, BUY COMPLETE , size = 5.00, price = 8.00

{datetime.datetime(2020, 1, 1, 0, 0): (100.0, 100.0),
 datetime.datetime(2020, 1, 2, 0, 0): (94.95, 99.95),
 datetime.datetime(2020, 1, 3, 0, 0): (84.85000000000001, 104.85000000000001),
 datetime.datetime(2020, 1, 4, 0, 0): (99.7, 114.7),
 datetime.datetime(2020, 1, 5, 0, 0): (119.5, 119.5),
 datetime.datetime(2020, 1, 6, 0, 0): (144.25, 119.25),
 datetime.datetime(2020, 1, 7, 0, 0): (173.95, 113.94999999999999),
 datetime.datetime(2020, 1, 8, 0, 0): (138.6, 103.6),
 datetime.datetime(2020, 1, 9, 0, 0): (98.19999999999999, 98.19999999999999)}

portfolio = vbt.Portfolio.from_orders(df.close, [np.nan] + size[:-1].tolist(), fees=0.01)
print(portfolio.cash(free=False))
print(portfolio.value())

2020-01-01    100.00
2020-01-02     94.95
2020-01-03     84.85
2020-01-04     99.70
2020-01-05    119.50
2020-01-06    144.25
2020-01-07    173.95
2020-01-08    138.60
2020-01-09     98.20
Name: close, dtype: float64
2020-01-01    100.00
2020-01-02     99.95
2020-01-03    104.85
2020-01-04    114.70
2020-01-05    119.50
2020-01-06    119.25
2020-01-07    113.95
2020-01-08    103.60
2020-01-09     98.20
Name: close, dtype: float64

strategy = bt_simulate(False)
strategy.analyzers.cashvalueanalyzer.get_analysis()

2020-01-02T00:00:00, BUY COMPLETE , size = 5.00, price = 1.00
2020-01-03T00:00:00, BUY COMPLETE , size = 5.00, price = 2.00
2020-01-04T00:00:00, SELL COMPLETE , size = -5.00, price = 3.00
2020-01-05T00:00:00, SELL COMPLETE , size = -5.00, price = 4.00
2020-01-06T00:00:00, SELL COMPLETE , size = -5.00, price = 5.00
2020-01-07T00:00:00, SELL COMPLETE , size = -5.00, price = 6.00
2020-01-08T00:00:00, BUY COMPLETE , size = 5.00, price = 7.00
2020-01-09T00:00:00, BUY COMPLETE , size = 5.00, price = 8.00

{datetime.datetime(2020, 1, 1, 0, 0): (100.0, 100.0),
 datetime.datetime(2020, 1, 2, 0, 0): (94.95, 99.95),
 datetime.datetime(2020, 1, 3, 0, 0): (84.85000000000001, 104.85000000000001),
 datetime.datetime(2020, 1, 4, 0, 0): (99.7, 114.7),
 datetime.datetime(2020, 1, 5, 0, 0): (119.5, 119.5),
 datetime.datetime(2020, 1, 6, 0, 0): (94.25, 119.25),
 datetime.datetime(2020, 1, 7, 0, 0): (63.95, 113.95),
 datetime.datetime(2020, 1, 8, 0, 0): (83.60000000000001, 103.60000000000001),
 datetime.datetime(2020, 1, 9, 0, 0): (98.2, 98.2)}

print(portfolio.cash(free=True))
print(portfolio.value())

2020-01-01    100.00
2020-01-02     94.95
2020-01-03     84.85
2020-01-04     99.70
2020-01-05    119.50
2020-01-06     94.25
2020-01-07     63.95
2020-01-08     83.60
2020-01-09     98.20
Name: close, dtype: float64
2020-01-01    100.00
2020-01-02     99.95
2020-01-03    104.85
2020-01-04    114.70
2020-01-05    119.50
2020-01-06    119.25
2020-01-07    113.95
2020-01-08    103.60
2020-01-09     98.20
Name: close, dtype: float64