Overall Statistics |
Total Trades 63 Average Win 0.00% Average Loss -0.02% Compounding Annual Return -6.924% Drawdown 0.500% Expectancy -0.964 Net Profit -0.498% Sharpe Ratio -16.159 Probabilistic Sharpe Ratio 0% Loss Rate 97% Win Rate 3% Profit-Loss Ratio 0.12 Alpha -0.054 Beta 0.002 Annual Standard Deviation 0.003 Annual Variance 0 Information Ratio -5.325 Tracking Error 0.246 Treynor Ratio -22.198 Total Fees $63.00 Estimated Strategy Capacity $0 Lowest Capacity Asset HD Y30X6E52BWBQ|HD R735QTJ8XC9X Portfolio Turnover 0.04% |
# region imports from AlgorithmImports import * import datetime as dt import ast import json from Selection.FundamentalUniverseSelectionModel import FundamentalUniverseSelectionModel from io import StringIO import pandas as pd from collections import defaultdict # endregion class AdaptableVioletJaguar(QCAlgorithm): def Initialize(self): #self.SetStartDate(2015, 1, 1) # Set Start Date self.SetStartDate(2022, 10, 1) self.SetEndDate(2022,11,1) self.SetCash(50000) # Set Strategy Cash self.tickers = ["CME", "AAPL", "MSFT", "HD", "MCD", "COST", "NVDA", "BBY"] self.AddUniverse(self.CoarseSelection, self.FineSelection) self.UniverseSettings.DataNormalizationMode = DataNormalizationMode.Raw self.symbolDataBySymbol = {} self.fineFundamentals = {} self.SAVE_EARNINGS = False self.loaded_earnings = {} self.slice = None if not self.SAVE_EARNINGS: self.SetEndDate(2023, 3, 11) earnings = json.loads(self.ObjectStore.Read("Earnings_dates")) temp_earnings = dict(earnings) for ticker, str_dates in earnings.items(): temp_earnings[ticker] = [dt.datetime.strptime(x, '%m/%d/%Y').date() for x in ast.literal_eval(str_dates)] self.loaded_earnings = dict(temp_earnings) self.Debug(temp_earnings) del temp_earnings del earnings elif self.SAVE_EARNINGS and self.ObjectStore.ContainsKey("Earnings_dates"): self.SetEndDate(2023, 1, 20) self.ObjectStore.Delete("Earnings_dates") self.spy = self.AddEquity("SPY", Resolution.Daily) self.Schedule.On(self.DateRules.EveryDay("SPY"), self.TimeRules.AfterMarketOpen("SPY", -10), self.CheckEarningsRules) self.Schedule.On(self.DateRules.EveryDay("SPY"), self.TimeRules.BeforeMarketClose("SPY", 60), self.UpdateEarningsObject) if not self.SAVE_EARNINGS: self.Schedule.On(self.DateRules.EveryDay("SPY"), self.TimeRules.BeforeMarketClose("SPY", 190), self.AddOptions) self.Schedule.On(self.DateRules.EveryDay("SPY"), self.TimeRules.BeforeMarketClose("SPY", 90), self.TradeOptions) self.Schedule.On(self.DateRules.EveryDay("SPY"), self.TimeRules.BeforeMarketClose("SPY", 150), self.CheckExits) self.Schedule.On(self.DateRules.EveryDay("SPY"), self.TimeRules.BeforeMarketClose("SPY", 30), self.AppendIV) def AppendIV(self): for symbol, symbolData in self.symbolDataBySymbol.items(): for contract, contract_object in symbolData.all_contracts.items(): contract_object.update_IV() def UpdateEarningsObject(self): if self.SAVE_EARNINGS: for symbol, symbolData in self.symbolDataBySymbol.items(): symbolData.update_earnings(self.fineFundamentals[symbol].EarningReports.FileDate) # Long Entry Criteria # at least 30 days since the last earnings report # over 30 days until the next earnings report # # Short Entry Criteria # less than 10 days until the next earnings report # # Short Exit Criteria # over 30 days since last earnings report # def CheckEarningsRules(self): if self.SAVE_EARNINGS: return cur_date = self.Time.date() for symbol, symbolData in self.symbolDataBySymbol.items(): symbol_earnings = symbolData.earnings_dates symbol_earnings_series = pd.Series(symbol_earnings) insertion_index = symbol_earnings_series.searchsorted(cur_date) ## Result is the index at which it will exit in the next list prev_earnings_date = symbol_earnings[insertion_index-1] next_earnings_date = symbol_earnings[insertion_index] ## WHEN RUNNING ALGORITHM, MAKE SURE self.EndDate IS LESS THAN THE LAST EARNINGS VALUE delta_prev = cur_date - prev_earnings_date delta_next = next_earnings_date - cur_date if delta_prev.days >= 30 and delta_next.days > 30: symbolData.long_entry_earnings = True else: symbolData.long_entry_earnings = False if delta_next.days < 10: symbolData.short_entry_earnings = True else: symbolData.short_entry_earnings = False if delta_prev.days >= 30: symbolData.k = True else: symbolData.short_exit_earnings = False if symbolData.long_entry_earnings or symbolData.short_entry_earnings: self.Debug(f"{insertion_index} {self.Time.date()} {symbol.Value} : long etr - {symbolData.long_entry_earnings}, short etr - {symbolData.short_entry_earnings}, short ext - {symbolData.short_exit_earnings}") def ContractFilter(self, symbol, min_strike, max_strike, min_expiry_days, max_expiry_days): contracts = self.OptionChainProvider.GetOptionContractList(symbol, self.Time.date()) if len(contracts) == 0 : return [] contract_list = [i for i in contracts if min_expiry_days < (i.ID.Date.date() - self.Time.date()).days < max_expiry_days] if len(contract_list) == 0: return [] min_strike_price = sorted(contract_list, key = lambda x: abs(x.ID.StrikePrice - min_strike))[0].ID.StrikePrice max_strike_price = sorted(contract_list, key = lambda x: abs(x.ID.StrikePrice - max_strike))[0].ID.StrikePrice strike_list = sorted(set([i.ID.StrikePrice for i in contract_list])) min_strike_rank = strike_list.index(min_strike_price) max_strike_rank = strike_list.index(max_strike_price) try: strikes = strike_list[min_strike_rank:max_strike_rank] except: strikes = strike_list filtered_contracts = [i for i in contract_list if i.ID.StrikePrice in strikes] return filtered_contracts def AddContract(self, slice, symbol, symbolData, position): security_price = self.Securities[symbol].Price lower_atr_strike = (symbolData.atr.Current.Value * 9) + security_price upper_atr_strike = (symbolData.atr.Current.Value * 11) + security_price cur_atr_strike = (symbolData.atr.Current.Value * 10) + security_price cur_atr = symbolData.atr.Current.Value if position == 'long': # self.Debug('long') # self.Debug(f"p: {security_price}, atr_l: {lower_atr_strike}, atr_u: {upper_atr_strike}, atr_c: {cur_atr_strike}, atr: {cur_atr}") filtered_contracts = self.ContractFilter(symbol, lower_atr_strike, upper_atr_strike, 300, 400) elif position == 'short': # self.Debug('short') # self.Debug(f"atr_l: {lower_atr_strike}, atr_u: {upper_atr_strike}, atr_c: {cur_atr_strike}, atr: {cur_atr}") filtered_contracts = self.ContractFilter(symbol, lower_atr_strike, upper_atr_strike, 20, 30) if len(filtered_contracts) == 0: return [] else: calls = [x for x in filtered_contracts if x.ID.OptionRight == OptionRight.Call] test = [x.ID.StrikePrice for x in calls] test2 = [x.ID.Date.date() for x in calls] self.Debug(f"Strikes: {test}") self.Debug(f"Expirys: {test2}") contracts = sorted(sorted(calls, key = lambda x: abs(self.Securities[symbol].Price- x.ID.StrikePrice)), key = lambda x: x.ID.Date, reverse=True) if len(contracts) == 0: return str() for contract in contracts: if contract not in self.symbolDataBySymbol[symbol].all_contracts: option = self.AddOptionContract(contract, Resolution.Hour) option.PriceModel = OptionPriceModels.BjerksundStensland() historical_IV = self.GetHistoricalIV(contract, symbol) self.symbolDataBySymbol[symbol].all_contracts[contract] = OptionContractIV(contract, historical_IV, position, symbol, self) def GetHistoricalIV(self, option_symbol, equity_symbol): Expiry = [] Strike = [] Type = [] Ticker = [] EndDate = [] IV = [] df_dict = defaultdict(list) requests = [] history_securities = [] look_back = 200 values = [] t_requests = [] option_security = None # FOR TESTING PURPOSES ############## # for security in self.Securities.Values: # all_symbol = security.Symbol # if all_symbol == option_symbol: # option_security = security # while len(values) == 0 or look_back > 100: # for subscription in option_security.Subscriptions: # t_requests.append(HistoryRequest(subscription, security.Exchange.Hours, self.Time.date()-timedelta(150), self.StartDate)) # t_history = self.History(t_requests) # for s in t_history: # for bar in s.Bars.Values: # values.append(bar) # break # look_back -= 1 ################### for security in self.Securities.Values: all_symbol = security.Symbol if all_symbol == option_symbol or all_symbol == equity_symbol: history_securities.append(security) ### RESOLUTION OF HISTORY REQUEST WILL BE WHAT THE SECURITY RESOLUTION IS for security in sorted(history_securities, key=lambda x: x.Type): for subscription in security.Subscriptions: requests.append(HistoryRequest(subscription, security.Exchange.Hours, self.Time.date()-timedelta(look_back), self.StartDate)) history = self.History(requests) ### SOMETIMES THERE IS NO OPTION HISTORICAL DATA. WHY ONLY SOMETIMES???? prev_date = None for s in history: underlying_price = None underlying_volatility = None for bar in s.QuoteBars.Values: self.Securities[bar.Symbol].SetMarketPrice(bar) for bar in s.Bars.Values: symbol = bar.Symbol security = self.Securities[symbol] security.SetMarketPrice(bar) if security.Type == SecurityType.Equity: underlying_price = security.Price continue ### WILL GET THE IV AT 10:00 (HOUR RESOLUTION)), MAYBE SHOULD TRY TO GET EOD? if len(df_dict['EndDate']) == 0 or bar.EndTime.date() > df_dict['EndDate'][-1].date(): contract = OptionContract.Create(symbol, symbol.Underlying, bar.EndTime, security, underlying_price) contract.LastPrice = bar.Close lastPrice = contract.LastPrice result = security.PriceModel.Evaluate(security, None, contract) IV.append(float(result.ImpliedVolatility)) df_dict['Expiry'].append(symbol.ID.Date) df_dict['Strike'].append(symbol.ID.StrikePrice) df_dict['EndDate'].append(bar.EndTime) df_dict['IV'].append(result.ImpliedVolatility) df = pd.DataFrame.from_dict(df_dict) # FOR DEBUGGING PURPOSES SPECIFICALLY FOR SETTING BREAKPOINT ### # if len(df.index) == 0: # self.Debug("") ### return df def AddOptions(self): for symbol, symbolData in self.symbolDataBySymbol.items(): if symbolData.long_call == None and symbolData.long_entry_earnings: self.AddContract(self.slice, symbol, symbolData, 'long') if symbolData.short_call == None and symbolData.long_entry_earnings: self.AddContract(self.slice, symbol, symbolData, 'short') def TradeOptions(self): if self.SAVE_EARNINGS: return for symbol, symbolData in self.symbolDataBySymbol.items(): for contract, contract_object in symbolData.all_contracts.items(): if symbolData.long_call == None and symbolData.long_entry_earnings: symbolData.make_trade_option_constraints(contract, 'long', self.slice.OptionChains.Values) if symbolData.short_call == None and symbolData.long_call != None and symbolData.short_entry_earnings: symbolData.make_trade_option_constraints(contract, 'short', self.slice.OptionChains.Values) def CheckExits(self): if self.SAVE_EARNINGS: return remove_contracts = {'long':False, 'short':False} pop_out_of_universe = [] for symbol, symbolData in self.symbolDataBySymbol.items(): for contract, contract_object in symbolData.all_contracts.items(): if symbolData.long_call == contract: # and self.Portfolio[contract].Invested: exit_df = contract_object.option_exit_data() pnl = exit_df['Pnl'] dt_expiry = exit_df['Dt_Expiry'] self.Debug(exit_df) if pnl >= 1 or pnl <= -0.5 or dt_expiry <= 90: self.Liquidate(contract) remove_contracts['long'] = True symbolData.long_call = None self.Debug('Liquidating Long Contract') elif symbolData.short_call == contract: # and self.Portfolio[contract].Invested: exit_df = contract_object.option_exit_data() pnl = exit_df['Pnl'] dt_expiry = exit_df['Dt_Expiry'] self.Debug(exit_df) if pnl >= 0.5 or self.Securities[symbol].Price > contract.ID.StrikePrice or symbolData.long_call == None: # NEED TO ADD 30 DAYS SINCE LAST EARNINGS REPORT self.Liquidate(contract) remove_contracts['short'] = True symbolData.short_call = None self.Debug('Liquidating Short Contract') ### NEED TO IMPLEMENT STRIKE OTM REMOVAL LOGIC ### IE. IF CONTRACT NO LONGER FITS STRIKE OTM 10xATR, THEN REMOVE elif contract_object.position == 'long': if contract_object.option_exit_data()['Dt_Expiry'] < 300: pop_out_of_universe.append(contract) self.RemoveOptionContract(contract) elif contract_object.position == 'short': if contract_object.option_exit_data()['Dt_Expiry'] < 20: pop_out_of_universe.append(contract) self.RemoveOptionContract(contract) for position, status in remove_contracts.items(): if status == True: remove = [] for symbol, symbolData in self.symbolDataBySymbol.items(): remove = [k for k,v in symbolData.all_contracts.items() if v.position == 'long'] for contract in remove: self.Debug(f"Removing : {contract} : {contract.SecurityType}") self.symbolDataBySymbol[contract.Underlying].all_contracts.pop(contract) self.RemoveOptionContract(contract) def OnData(self, data: Slice): self.slice = data def OnSecuritiesChanged(self, changes): for security in changes.AddedSecurities: if security.Symbol.Value == "SPY" or security.Symbol.SecurityType != SecurityType.Equity: continue # NEED TO LOOK INTO THE IMPACT OF THIS ON IV, SPECIFICALLY THE RESOLUTION self.Securities[security.Symbol].VolatilityModel = StandardDeviationOfReturnsVolatilityModel(30)#, Resolution.Daily) # WITH SOME SECURITIES SETTING TO RAW DIDNT WORK, FIXED BY JUST SETTING UNIVERSE TO RAW # security.SetDataNormalizationMode(DataNormalizationMode.Raw) atr = self.ATR(security.Symbol, 20, MovingAverageType.Simple, Resolution.Daily) history = self.History(security.Symbol, 20, Resolution.Daily) for bar in history.itertuples(): tradebar = TradeBar(bar.Index[1], security.Symbol, bar.open, bar.high, bar.low, bar.close, bar.volume) atr.Update(tradebar) symbolData = SymbolData(self, security, atr) if not self.SAVE_EARNINGS: symbolData.earnings_dates = self.loaded_earnings[security.Symbol.Value] self.symbolDataBySymbol[security.Symbol] = symbolData for security in changes.RemovedSecurities: if security.Symbol.SecurityType == SecurityType.Equity and self.symbolDataBySymbol[security.Symbol].remove_symbol == True: self.symbolDataBySymbol.pop(security.Symbol) elif security.Symbol.SecurityType == SecurityType.Option: ### This if statement is there to prevent a bug where key i not found if security.Symbol in self.symbolDataBySymbol[security.Underlying.Symbol].all_contracts: self.symbolDataBySymbol[security.Underlying.Symbol].all_contracts.pop(security.Symbol) for symbol in self.Securities.Keys: if symbol.SecurityType == SecurityType.Option and symbol.Underlying == security.Symbol: self.RemoveSecurity(symbol) def CoarseSelection(self, coarse): filteredCoarse = [x.Symbol for x in coarse if x.Symbol.Value in self.tickers and x.HasFundamentalData] return filteredCoarse def FineSelection(self, fine): for x in fine: self.fineFundamentals[x.Symbol] = x return [x.Symbol for x in fine] def OnEndOfAlgorithm(self): if self.SAVE_EARNINGS: earnings_dict = {} for symbol, symbolData in self.symbolDataBySymbol.items(): self.Debug(f"{symbol.Value} : {symbolData.earnings_dates}") string_earnings = [t.strftime('%m/%d/%Y') for t in symbolData.earnings_dates] earnings_dict[str(symbol.Value)] = str(string_earnings) dump = json.dumps(earnings_dict) self.ObjectStore.Save("Earnings_dates", dump) self.Debug(earnings_dict) else: for symbol, symbolData in self.symbolDataBySymbol.items(): self.Debug(symbolData.earnings_dates) class SymbolData: def __init__(self, algo, security, atr): self.algo = algo self.security = security self.symbol = security.Symbol self.earnings_dates = [] self.long_entry_earnings = False self.short_entry_earnings = False self.short_exit_earnings = False self.long_call = None self.short_call = None self.atr = atr self.all_contracts = {} self.remove_symbol = False def update_earnings(self, date): if date.date() not in self.earnings_dates: self.earnings_dates.append(date.date()) ## RUN THIS EVERYONE MORNING def check_option_constraints(self): pass def make_trade_option_constraints(self, contract, position, optionChain): call = None for chain in optionChain: for x in chain: if x.Symbol == contract: call = x self.algo.Debug(f"Looking at {call} : {call.Expiry} : {call.Strike} - {position}") # SOMETIMES A CONTRACT HAS IV SOMETIMES IT DOESNT. NEED TO LOOK INTO WHY. # CANT FIGURE OUT WHY VIA RESEARCH NOTEBOOK. NEED TO CREATE SEPARATE ALGO TO TEST try: cur_IV = float(call.ImpliedVolatility) except: self.algo.Debug(f"No IV: {contract}") return if len(self.all_contracts[contract].historical_IV.index) == 0: hist_iv_df = self.algo.GetHistoricalIV(contract, self.symbol) else: hist_iv_df = self.all_contracts[contract].historical_IV # SOMETIMES A CONTRACT HAS HISTORICAL IV SOMETIMES IT DOESNT. NEED TO LOOK INTO WHY. # CANT FIGURE OUT WHY VIA RESEARCH NOTEBOOK. NEED TO CREATE SEPARATE ALGO TO TEST try: hist_iv = list(hist_iv_df.IV) except: self.algo.Debug(f"No HIST IV: {self.algo.Time.date()} - {contract} ")#- {contract.Expiry}") return hist_iv = hist_iv[:100] IV_Percentile = sum(cur_IV < i for i in hist_iv) / 100 self.algo.Debug(f"IV_Percentile : {IV_Percentile}") if position == 'long' and IV_Percentile <= 0.1 and IV_Percentile > 0: self.algo.Buy(contract,1) self.long_call = contract self.algo.Debug(f"Trade on : {position} : {self.algo.Time} : {call.Strike} : {call.Expiry} - {contract.Underlying}") elif position == 'short' and IV_Percentile > 0.9: self.algo.Sell(contract,1) self.short_call = contract self.algo.Debug(f"Trade on : {position} : {self.algo.Time} : {call.Strike} : {call.Expiry} - {contract.Underlying}") class OptionContractIV: def __init__(self, symbol, historical_IV, position, ticker, algo): self.symbol = symbol self.historical_IV = historical_IV self.position = position self.ticker = ticker self.algo = algo def update_IV(self): call = None for chain in self.algo.slice.OptionChains.Values: for x in chain: if x.Symbol == self.symbol: call = x if call == None: return df_dict = defaultdict(list) df_dict['Expiry'].append(self.symbol.ID.Date) df_dict['Strike'].append(self.symbol.ID.StrikePrice) df_dict['EndDate'].append(self.algo.Time) df_dict['IV'].append(call.ImpliedVolatility) df = pd.DataFrame.from_dict(df_dict) self.historical_IV = pd.concat([self.historical_IV, df], ignore_index=True) def option_exit_data(self): exits = {} exits['Pnl'] = self.algo.Portfolio[self.symbol].UnrealizedProfitPercent delta = (self.symbol.ID.Date.date() - self.algo.Time.date()) exits['Dt_Expiry'] = delta.days return exits