| Overall Statistics |
|
Total Trades 68 Average Win 6.93% Average Loss -0.83% Compounding Annual Return 383.712% Drawdown 29.500% Expectancy 5.328 Net Profit 449.830% Sharpe Ratio 7.486 Probabilistic Sharpe Ratio 94.664% Loss Rate 32% Win Rate 68% Profit-Loss Ratio 8.38 Alpha 3.302 Beta 2.702 Annual Standard Deviation 0.689 Annual Variance 0.475 Information Ratio 8.128 Tracking Error 0.55 Treynor Ratio 1.91 Total Fees $331.48 Estimated Strategy Capacity $70000000.00 |
class StatisticalArbitrage(QCAlgorithm):
def Initialize(self):
self.SetStartDate(2020, 4, 1)
self.SetCash(100000)
data = self.AddEquity("SPY", Resolution.Minute)
data.SetLeverage(20)
self.symbol = data.Symbol
option = self.AddOption("SPY", Resolution.Minute)
option.SetFilter(-20, 20, 25, 35)
self.last_day = -1
def OnData(self,slice):
# Check once a day.
if self.Time.day == self.last_day:
return
self.last_day = self.Time.day
for i in slice.OptionChains:
chains = i.Value
if not self.Portfolio.Invested:
# divide option chains into call and put options
calls = list(filter(lambda x: x.Right == OptionRight.Call, chains))
puts = list(filter(lambda x: x.Right == OptionRight.Put, chains))
# if lists are empty return
if not calls or not puts: return
underlying_price = self.Securities[self.symbol].Price
expiries = [i.Expiry for i in puts]
# determine expiration date nearly one month
expiry = min(expiries, key=lambda x: abs((x.date()-self.Time.date()).days-30))
strikes = [i.Strike for i in puts]
# determine at-the-money strike
strike = min(strikes, key=lambda x: abs(x-underlying_price))
# determine 15% out-of-the-money strike
otm_strike = min(strikes, key = lambda x:abs(x - float(0.85) * underlying_price))
atm_call = [i for i in calls if i.Expiry == expiry and i.Strike == strike]
atm_put = [i for i in puts if i.Expiry == expiry and i.Strike == strike]
otm_put = [i for i in puts if i.Expiry == expiry and i.Strike == otm_strike]
if atm_call and atm_put and otm_put:
options_q = int(self.Portfolio.MarginRemaining / (underlying_price * 100))
# Set max leverage.
self.Securities[atm_call[0].Symbol].MarginModel = BuyingPowerModel(20)
self.Securities[atm_put[0].Symbol].MarginModel = BuyingPowerModel(20)
self.Securities[otm_put[0].Symbol].MarginModel = BuyingPowerModel(20)
# sell at-the-money straddle
self.Sell(atm_call[0].Symbol, options_q)
self.Sell(atm_put[0].Symbol, options_q)
# buy 15% out-of-the-money put
self.Buy(otm_put[0].Symbol, options_q)
# buy index.
self.SetHoldings(self.symbol, 5)
invested = [x.Key for x in self.Portfolio if x.Value.Invested]
if len(invested) == 1:
self.Liquidate(self.symbol)class StatisticalArbitrage(QCAlgorithm):
def Initialize(self):
self.SetStartDate(2020, 4, 1)
self.SetCash(100000)
data = self.AddEquity("SPY", Resolution.Minute)
data.SetLeverage(20)
self.symbol = data.Symbol
option = self.AddOption("SPY", Resolution.Minute)
option.SetFilter(-20, 20, 25, 35)
self.last_day = -1
def OnData(self,slice):
# Check once a day.
if self.Time.day == self.last_day:
return
self.last_day = self.Time.day
for i in slice.OptionChains:
chains = i.Value
if not self.Portfolio.Invested:
# divide option chains into call and put options
calls = list(filter(lambda x: x.Right == OptionRight.Call, chains))
puts = list(filter(lambda x: x.Right == OptionRight.Put, chains))
# if lists are empty return
if not calls or not puts: return
underlying_price = self.Securities[self.symbol].Price
expiries = [i.Expiry for i in puts]
# determine expiration date nearly one month
expiry = min(expiries, key=lambda x: abs((x.date()-self.Time.date()).days-30))
strikes = [i.Strike for i in puts]
# determine at-the-money strike
strike = min(strikes, key=lambda x: abs(x-underlying_price))
# determine 15% out-of-the-money strike
otm_strike = min(strikes, key = lambda x:abs(x - float(0.85) * underlying_price))
atm_call = [i for i in calls if i.Expiry == expiry and i.Strike == strike]
atm_put = [i for i in puts if i.Expiry == expiry and i.Strike == strike]
otm_put = [i for i in puts if i.Expiry == expiry and i.Strike == otm_strike]
if atm_call and atm_put and otm_put:
options_q = int(self.Portfolio.MarginRemaining / (underlying_price * 100))
# Set max leverage.
self.Securities[atm_call[0].Symbol].MarginModel = BuyingPowerModel(20)
self.Securities[atm_put[0].Symbol].MarginModel = BuyingPowerModel(20)
self.Securities[otm_put[0].Symbol].MarginModel = BuyingPowerModel(20)
# sell at-the-money straddle
self.Sell(atm_call[0].Symbol, options_q)
self.Sell(atm_put[0].Symbol, options_q)
# buy 15% out-of-the-money put
self.Buy(otm_put[0].Symbol, options_q)
# buy index.
self.SetHoldings(self.symbol, 5)
invested = [x.Key for x in self.Portfolio if x.Value.Invested]
if len(invested) == 1:
self.Liquidate(self.symbol)