Overall Statistics Total Trades 47 Average Win 8.41% Average Loss -2.14% Compounding Annual Return 10.305% Drawdown 19.500% Expectancy 2.851 Net Profit 269.970% Sharpe Ratio 0.945 Loss Rate 22% Win Rate 78% Profit-Loss Ratio 3.92 Alpha 0.017 Beta 4.366 Annual Standard Deviation 0.11 Annual Variance 0.012 Information Ratio 0.764 Tracking Error 0.11 Treynor Ratio 0.024 Total Fees \$433.22
from datetime import datetime,timedelta
import pandas as pd
import numpy as np

class PairedSwitching(QCAlgorithm):

def Initialize(self):
self.SetStartDate(2005,3,15)
self.SetEndDate(2018,7,15)
self.SetCash(100000)
#we select two etfs that are negatively correlated; equity and bond etfs
self.months = -1
#monthly scheduled event but rebalancing will run on quarterly basis
self.Schedule.On(self.DateRules.MonthStart("SPY"), self.TimeRules.AfterMarketOpen("SPY", 1), self.Rebalance)

def Rebalance(self):
self.months +=1
if(self.months%3==0):
#retrieves prices from 90 days ago
history_call = self.History(self.Securities.Keys,timedelta(days=90))
if not history_call.empty:
first_bars = history_call.loc[self.first.Symbol.Value]
last_p1 = first_bars["close"].iloc[0]
second_bars = history_call.loc[self.second.Symbol.Value]
last_p2 = second_bars["close"].iloc[0]
# calculates performance of funds over the prior quarter
first_performance = (float(self.Securities[self.first.Symbol].Price) - float(last_p1))/(float(self.Securities[self.first.Symbol].Price))
second_performance = (float(self.Securities[self.second.Symbol].Price) - float(last_p2))/(float(self.Securities[self.second.Symbol].Price))
#buys the fund that has the higher return during the period
if(first_performance > second_performance):
if(self.Securities[self.second.Symbol].Invested==True):
self.Liquidate(self.second.Symbol)
self.SetHoldings(self.first.Symbol,1)
else:
if(self.Securities[self.first.Symbol].Invested==True):
self.Liquidate(self.first.Symbol)
self.SetHoldings(self.second.Symbol,1)

def OnData(self, data):
pass