Overall Statistics
Total Trades
986
Average Win
0.30%
Average Loss
-0.10%
Compounding Annual Return
11.321%
Drawdown
17.500%
Expectancy
1.224
Net Profit
80.561%
Sharpe Ratio
0.875
Probabilistic Sharpe Ratio
33.659%
Loss Rate
43%
Win Rate
57%
Profit-Loss Ratio
2.91
Alpha
0.093
Beta
0.058
Annual Standard Deviation
0.113
Annual Variance
0.013
Information Ratio
-0.018
Tracking Error
0.196
Treynor Ratio
1.723
Total Fees
$1171.51
import numpy as np


class TransdimensionalOptimizedAtmosphericScrubbers(QCAlgorithm):

    def Initialize(self):
        self.SetStartDate(2015, 4, 29)  # Set Start Date
        self.SetCash(100000)  # Set Strategy Cash

        res = Resolution.Daily

        self.STOCKS = [self.AddEquity('QQQ', res).Symbol]
        self.BONDS = [self.AddEquity(ticker, res).Symbol for ticker in ['TLT', 'IEF']]

        self.XLI = self.AddEquity('XLI', res).Symbol
        self.XLU = self.AddEquity('XLU', res).Symbol
        self.UUP = self.AddEquity('UUP', res).Symbol
        self.MKT = self.STOCKS[0]

        self.VOLA = 126;
        self.BULL = 1;
        self.COUNT = 0;
        self.OUT_DAY = 0;
        self.RET_INITIAL = 80;
        self.LEV = 1.00;

        self.Schedule.On(self.DateRules.EveryDay(), self.TimeRules.AfterMarketOpen('QQQ', 140), self.daily_check)

    def daily_check(self):
        vola = self.History([self.MKT], self.VOLA + 1, Resolution.Daily).loc[self.MKT][
                   'close'].pct_change().std() * np.sqrt(252)
        WAIT_DAYS = int(vola * self.RET_INITIAL)
        RET = int((1.0 - vola) * self.RET_INITIAL)

        P = self.History([self.XLI, self.XLU, self.UUP], RET + 2, Resolution.Daily)['close'].unstack(level=0).iloc[:-1].dropna()
        if (len(P.columns) < 2):
            return
        ratio = (P[self.XLI].iloc[-1] / P[self.XLI].iloc[0]) / (P[self.XLU].iloc[-1] / P[self.XLU].iloc[0])

        exit = ratio < 1.0
        if exit:
            self.BULL = 0;
            self.OUT_DAY = self.COUNT;
        elif (self.COUNT >= self.OUT_DAY + WAIT_DAYS):
            self.BULL = 1
        self.COUNT += 1

        wt_stk = self.LEV if self.BULL else 0;
        wt_bnd = 0 if self.BULL else self.LEV;

        wt = {}

        for sec in self.STOCKS:
            wt[sec] = wt_stk / len(self.STOCKS);

        for sec in self.BONDS:
            wt[sec] = wt_bnd / len(self.BONDS)

        for sec, weight in wt.items():
            self.SetHoldings(sec, weight)