Overall Statistics
Total Trades
567
Average Win
3.06%
Average Loss
-2.32%
Compounding Annual Return
18.686%
Drawdown
41.400%
Expectancy
0.376
Net Profit
911.538%
Sharpe Ratio
0.747
Probabilistic Sharpe Ratio
10.135%
Loss Rate
41%
Win Rate
59%
Profit-Loss Ratio
1.32
Alpha
0.151
Beta
0.365
Annual Standard Deviation
0.253
Annual Variance
0.064
Information Ratio
0.314
Tracking Error
0.271
Treynor Ratio
0.518
Total Fees
$3022.31
Estimated Strategy Capacity
$19000000.00
Lowest Capacity Asset
TWTR VLE97YS7S57P
"""
SEL(stock selection part)
Valuation Rockets
Based on the 'Momentum Strategy with Market Cap and EV/EBITDA' strategy introduced by Jing Wu, 6 Feb 2018
adapted and recoded by Jack Simonson, Goldie Yalamanchi, Vladimir, Peter Guenther, Leandro Maia, Mark Hatlan, and Simone Pantaleoni

https://www.quantconnect.com/forum/discussion/3377/momentum-strategy-with-market-cap-and-ev-ebitda/p1
https://www.quantconnect.com/forum/discussion/9678/quality-companies-in-an-uptrend/p1
https://www.quantconnect.com/forum/discussion/9632/amazing-returns-superior-stock-selection-strategy-superior-in-amp-out-strategy/p1

I/O(in & out part)
Option 1: The In & Out algo
Based on the 'In & Out' strategy introduced by Peter Guenther, 4 Oct 2020
expanded/inspired by Tentor Testivis, Dan Whitnable (Quantopian), Vladimir, Thomas Chang, 
Mateusz Pulka, Derek Melchin (QuantConnect), Nathan Swenson, Goldie Yalamanchi, and Sudip Sil

https://www.quantopian.com/posts/new-strategy-in-and-out
https://www.quantconnect.com/forum/discussion/9597/the-in-amp-out-strategy-continued-from-quantopian/p1

Option 2: The Distilled Bear in & out algo
based on Dan Whitnable's 22 Oct 2020 algo on Quantopian. 
Dan's original notes:
"This is based on Peter Guenther great “In & Out” algo.
Included Tentor Testivis recommendation to use volatility adaptive calculation of WAIT_DAYS and RET.
Included Vladimir's ideas to eliminate fixed constants
Help from Thomas Chang"

https://www.quantopian.com/posts/new-strategy-in-and-out
https://www.quantconnect.com/forum/discussion/9597/the-in-amp-out-strategy-continued-from-quantopian/
"""

from QuantConnect.Data.UniverseSelection import *
import math
import numpy as np
import pandas as pd
import scipy as sp

class ValuationRockets_inout(QCAlgorithm):

    def Initialize(self):

        self.SetStartDate(2008, 1, 1)  #Set Start Date
        #self.SetEndDate(2010, 12, 31)  #Set End Date
        self.cap = 100000
        self.SetCash(self.cap)
        
        res = Resolution.Hour
        
        # Holdings
        ### 'Out' holdings and weights
        self.BND1 = self.AddEquity('TLT', res).Symbol #TLT; TMF for 3xlev
        self.quantity = {self.BND1: 0}
        
        # Choose in & out algo
        self.go_inout_vs_dbear = 1 # 1=In&Out, 0=DistilledBear
        
        ##### In & Out parameters #####
        # Feed-in constants
        self.INI_WAIT_DAYS = 15  # out for 3 trading weeks
        self.wait_days = self.INI_WAIT_DAYS
        
        # Market and list of signals based on ETFs
        self.MRKT = self.AddEquity('SPY', res).Symbol  # market
        self.PRDC = self.AddEquity('XLI', res).Symbol  # production (industrials)
        self.METL = self.AddEquity('DBB', res).Symbol  # input prices (metals)
        self.NRES = self.AddEquity('IGE', res).Symbol  # input prices (natural res)
        self.DEBT = self.AddEquity('SHY', res).Symbol  # cost of debt (bond yield)
        self.USDX = self.AddEquity('UUP', res).Symbol  # safe haven (USD)
        self.GOLD = self.AddEquity('GLD', res).Symbol  # gold
        self.SLVA = self.AddEquity('SLV', res).Symbol  # vs silver
        #self.INFL = self.AddEquity('RINF', res).Symbol  # disambiguate GPLD/SLVA pair via inflaction expectations
        self.TIPS = self.AddEquity('TIP', res).Symbol  # disambiguate GPLD/SLVA pair via inflaction expectations; Treasury Yield = TIPS Yield + Expected Inflation
        self.UTIL = self.AddEquity('XLU', res).Symbol  # utilities
        self.INDU = self.PRDC  # vs industrials
        self.SHCU = self.AddEquity('FXF', res).Symbol  # safe haven currency (CHF)
        self.RICU = self.AddEquity('FXA', res).Symbol  # vs risk currency (AUD)

        self.FORPAIRS = [self.GOLD, self.SLVA, self.UTIL, self.SHCU, self.RICU, self.TIPS] #self.INFL
        self.SIGNALS = [self.PRDC, self.METL, self.NRES, self.DEBT, self.USDX]
        self.pairlist = ['G_S', 'U_I', 'C_A']
        
        # Initialize variables
        ## 'In'/'out' indicator
        self.be_in = 1 #-1 #initially, set to an arbitrary value different from 1 (in) and 0 (out)
        self.be_in_prior = 0 #-1 #initially, set to an arbitrary value different from 1 (in) and 0 (out)
        ## Day count variables
        self.dcount = 0  # count of total days since start
        self.outday = (-self.INI_WAIT_DAYS+1)  # setting ensures universe updating at algo start
        ## Flexi wait days
        self.WDadjvar = self.INI_WAIT_DAYS
        self.adjwaitdays = self.INI_WAIT_DAYS
        ## For inflation gauge
        self.debt1st = []
        self.tips1st = []
        
        ##### Distilled Bear parameters (note: some signals shared with In & Out) #####
        self.DISTILLED_BEAR = 1 #-1
        self.VOLA_LOOKBACK = 126
        self.WAITD_CONSTANT = 85
        
        # set a warm-up period to initialize the indicator
        self.SetWarmUp(timedelta(350))
        
        ##### Valuation Rockets parameters #####
        self.UniverseSettings.Resolution = res
        self.AddUniverse(self.UniverseCoarseFilter, self.UniverseFundamentalsFilter)
        self.num_coarse = 100
        self.num_screener = 20
        self.num_stocks = 5
        self.formation_days = 126
        self.lowmom = False
        self.data = {}
        self.setrebalancefreq = 60 # X days, update universe and momentum calculation
        self.updatefinefilter = 0
        self.symbols = None
        self.reb_count = 0
        
        self.Schedule.On(
            self.DateRules.EveryDay(),
            self.TimeRules.AfterMarketOpen('SPY', 30),
            self.rebalance_when_out_of_the_market)
        
        self.Schedule.On(
            self.DateRules.EveryDay(), 
            self.TimeRules.BeforeMarketClose('SPY', 0), 
            self.record_vars)  
        
        # Benchmarks
        self.QQQ = self.AddEquity('QQQ', res).Symbol
        self.benchmarks = []
        self.year = self.Time.year #for resetting benchmarks annually if applicable
        
        # Setup daily consolidation
        symbols = [self.MRKT] + self.SIGNALS + self.FORPAIRS + [self.QQQ]
        for symbol in symbols:
            self.consolidator = TradeBarConsolidator(timedelta(days=1))
            self.consolidator.DataConsolidated += self.consolidation_handler
            self.SubscriptionManager.AddConsolidator(symbol, self.consolidator)
        
        # Warm up history
        if self.go_inout_vs_dbear==1: self.lookback = 252
        if self.go_inout_vs_dbear==0: self.lookback = 126
        self.history = self.History(symbols, self.lookback, Resolution.Daily)
        if self.history.empty or 'close' not in self.history.columns:
            return
        self.history = self.history['close'].unstack(level=0).dropna()
        
    def UniverseCoarseFilter(self, coarse):
        if not (((self.dcount-self.reb_count)==self.setrebalancefreq) or (self.dcount == self.outday + self.adjwaitdays - 1)):
            self.updatefinefilter = 0
            return Universe.Unchanged
        
        self.updatefinefilter = 1
            
        # drop stocks which have no fundamental data or have too low prices
        selected = [x for x in coarse if (x.HasFundamentalData) and (float(x.Price) > 5)]
        # rank the stocks by dollar volume 
        filtered = sorted(selected, key=lambda x: x.DollarVolume, reverse=True)
        return [x.Symbol for x in filtered[:self.num_coarse]]
        
        
    def UniverseFundamentalsFilter(self, fundamental):
        if self.updatefinefilter == 0:
            return Universe.Unchanged
            
        filtered_fundamental = [x for x in fundamental if (x.ValuationRatios.EVToEBITDA > 0) 
                                        and (x.EarningReports.BasicAverageShares.ThreeMonths > 0) 
                                        and float(x.EarningReports.BasicAverageShares.ThreeMonths) * x.Price > 2e9
                                        and x.SecurityReference.IsPrimaryShare
                                        and x.SecurityReference.SecurityType == "ST00000001"
                                        and x.SecurityReference.IsDepositaryReceipt == 0
                                        and x.CompanyReference.IsLimitedPartnership == 0]

        top = sorted(filtered_fundamental, key = lambda x: x.ValuationRatios.EVToEBITDA, reverse=True)[:self.num_screener]
        self.symbols = [x.Symbol for x in top]
        self.updatefinefilter = 0
        self.reb_count = self.dcount
        return self.symbols

    
    def OnSecuritiesChanged(self, changes):
        
        addedSymbols = []
        for security in changes.AddedSecurities:
            addedSymbols.append(security.Symbol)
            if security.Symbol not in self.data:
                self.data[security.Symbol] = SymbolData(security.Symbol, self.formation_days, self)
   
        if len(addedSymbols) > 0:
            history = self.History(addedSymbols, 1 + self.formation_days, Resolution.Daily).loc[addedSymbols]
            for symbol in addedSymbols:
                try:
                    self.data[symbol].Warmup(history.loc[symbol])
                except:
                    self.Debug(str(symbol))
                    continue
    
    def consolidation_handler(self, sender, consolidated):
        self.history.loc[consolidated.EndTime, consolidated.Symbol] = consolidated.Close
        self.history = self.history.iloc[-self.lookback:]
        if self.go_inout_vs_dbear==1: self.update_history_shift()
    
    def update_history_shift(self):
        self.history_shift = self.history.rolling(11, center=True).mean().shift(60)
        
    def derive_vola_waitdays(self):
        volatility = 0.6 * np.log1p(self.history[[self.MRKT]].pct_change()).std() * np.sqrt(252)
        wait_days = int(volatility * self.WAITD_CONSTANT)
        returns_lookback = int((1.0 - volatility) * self.WAITD_CONSTANT)
        return wait_days, returns_lookback
    
    def signalcheck_inout(self):
        ##### In & Out signal check logic #####
        
        # Returns sample to detect extreme observations
        returns_sample = (self.history / self.history_shift - 1)
        # Reverse code USDX: sort largest changes to bottom
        returns_sample[self.USDX] = returns_sample[self.USDX] * (-1)
        # For pairs, take returns differential, reverse coded
        returns_sample['G_S'] = -(returns_sample[self.GOLD] - returns_sample[self.SLVA])
        returns_sample['U_I'] = -(returns_sample[self.UTIL] - returns_sample[self.INDU])
        returns_sample['C_A'] = -(returns_sample[self.SHCU] - returns_sample[self.RICU])   

        # Extreme observations; statist. significance = 1%
        pctl_b = np.mean(returns_sample, axis=0)-2.58*np.std(returns_sample, axis=0)
        extreme_b = returns_sample.iloc[-1] < pctl_b
        
        # Re-assess/disambiguate double-edged signals
        if self.dcount==0:
            self.debt1st = self.history[self.DEBT]
            self.tips1st = self.history[self.TIPS]
        self.history['INFL'] = (self.history[self.DEBT]/self.debt1st - self.history[self.TIPS]/self.tips1st)
        median = np.nanmedian(self.history, axis=0)
        abovemedian = self.history.iloc[-1] > median
        ### Interest rate expectations (cost of debt) may increase because the economic outlook improves (showing in rising input prices) = actually not a negative signal
        extreme_b.loc[[self.DEBT]] = np.where((extreme_b.loc[[self.DEBT]].any()) & (abovemedian[[self.METL, self.NRES]].any()), False, extreme_b.loc[[self.DEBT]])
        ### GOLD/SLVA differential may increase due to inflation expectations which actually suggest an economic improvement = actually not a negative signal
        extreme_b.loc['G_S'] = np.where((extreme_b.loc[['G_S']].any()) & (abovemedian.loc[['INFL']].any()), False, extreme_b.loc['G_S'])

        # Determine waitdays empirically via safe haven excess returns, 50% decay
        self.WDadjvar = int(
            max(0.50 * self.WDadjvar,
                self.INI_WAIT_DAYS * max(1,
                                         np.where((returns_sample[self.GOLD].iloc[-1]>0) & (returns_sample[self.SLVA].iloc[-1]<0) & (returns_sample[self.SLVA].iloc[-2]>0), self.INI_WAIT_DAYS, 1),
                                         np.where((returns_sample[self.UTIL].iloc[-1]>0) & (returns_sample[self.INDU].iloc[-1]<0) & (returns_sample[self.INDU].iloc[-2]>0), self.INI_WAIT_DAYS, 1),
                                         np.where((returns_sample[self.SHCU].iloc[-1]>0) & (returns_sample[self.RICU].iloc[-1]<0) & (returns_sample[self.RICU].iloc[-2]>0), self.INI_WAIT_DAYS, 1)
                                         ))
        )
        self.adjwaitdays = min(60, self.WDadjvar)
        
        return (extreme_b[self.SIGNALS + self.pairlist]).any()
    
    def signalcheck_dbear(self):
        ##### Distilled Bear signal check logic #####
        
        self.adjwaitdays, returns_lookback = self.derive_vola_waitdays()
        
        ## Check for Bears
        returns = self.history.pct_change(returns_lookback).iloc[-1]
    
        silver_returns = returns[self.SLVA]
        gold_returns = returns[self.GOLD]
        industrials_returns = returns[self.INDU]
        utilities_returns = returns[self.UTIL]
        metals_returns = returns[self.METL]
        dollar_returns = returns[self.USDX]
        
        DISTILLED_BEAR = (((gold_returns > silver_returns) and
                       (utilities_returns > industrials_returns)) and 
                       (metals_returns < dollar_returns)
                       )
        
        return DISTILLED_BEAR
    
        
    def rebalance_when_out_of_the_market(self):
        
        if self.go_inout_vs_dbear==1: out_signal = self.signalcheck_inout()
        if self.go_inout_vs_dbear==0: out_signal = self.signalcheck_dbear()
            
        ##### Determine whether 'in' or 'out' of the market. Perform out trading if applicable #####
        
        if out_signal:
            self.be_in = False
            self.outday = self.dcount
            
            if self.quantity[self.BND1] == 0:
                for symbol in self.quantity.copy().keys():
                    if symbol == self.BND1: continue
                    self.Order(symbol, - self.quantity[symbol])
                    self.Debug([str(self.Time), str(symbol), str(-self.quantity[symbol])])
                    del self.quantity[symbol]
                quantity = self.Portfolio.TotalPortfolioValue / self.Securities[self.BND1].Close
                self.quantity[self.BND1] = math.floor(quantity)
                self.Order(self.BND1, self.quantity[self.BND1])
                self.Debug([str(self.Time), str(self.BND1), str(self.quantity[self.BND1])])
        
        if (self.dcount >= self.outday + self.adjwaitdays):
            self.be_in = True
        
        # Update stock ranking/holdings, when swithing from 'out' to 'in' plus every X days when 'in' (set rebalance frequency)
        if (self.be_in and not self.be_in_prior) or (self.be_in and (self.dcount==self.reb_count)):
            self.rebalance()
                
        self.be_in_prior = self.be_in
        self.dcount += 1


    def rebalance(self):
            
        if self.symbols is None: return
        chosen_df = self.calc_return(self.symbols)
        chosen_df = chosen_df.iloc[:self.num_stocks]
        
        if self.quantity[self.BND1] > 0:
            self.Order(self.BND1, - self.quantity[self.BND1])
            self.Debug([str(self.Time), str(self.BND1), str(-self.quantity[self.BND1])])
            self.quantity[self.BND1] = 0
            
        weight = 1 / self.num_stocks
        for symbol in self.quantity.copy().keys():
            if symbol == self.BND1: continue
            if not self.CurrentSlice.ContainsKey(symbol) or self.CurrentSlice[symbol] is None:
                continue
            if symbol not in chosen_df.index:
                self.Order(symbol, - self.quantity[symbol])
                self.Debug([str(self.Time), str(symbol), str(-self.quantity[symbol])])
                del self.quantity[symbol]
            else:
                quantity = self.Portfolio.TotalPortfolioValue * weight / self.Securities[symbol].Close
                if math.floor(quantity) != self.quantity[symbol]:
                    self.Order(symbol, math.floor(quantity) - self.quantity[symbol])
                    self.Debug([str(self.Time), str(symbol), str(math.floor(quantity) -self.quantity[symbol])])
                    self.quantity[symbol] = math.floor(quantity)
        
        for symbol in chosen_df.index:
            if not self.CurrentSlice.ContainsKey(symbol) or self.CurrentSlice[symbol] is None:
                continue
            if symbol not in self.quantity.keys():
                quantity = self.Portfolio.TotalPortfolioValue * weight / self.Securities[symbol].Close
                self.quantity[symbol] = math.floor(quantity)
                self.Order(symbol, self.quantity[symbol])
                self.Debug([str(self.Time), str(symbol), str(self.quantity[symbol])])
 
        
    def calc_return(self, stocks):
        
        ret = {}
        for symbol in stocks:
            try:
                ret[symbol] = self.data[symbol].Roc.Current.Value
            except:
                self.Debug(str(symbol))
                continue
            
        df_ret = pd.DataFrame.from_dict(ret, orient='index')
        df_ret.columns = ['return']
        sort_return = df_ret.sort_values(by = ['return'], ascending = self.lowmom)
        
        return sort_return
    
        
    def record_vars(self): 
        
        if self.dcount==1: self.benchmarks = [self.history[self.MRKT].iloc[-2], self.Portfolio.TotalPortfolioValue, self.history[self.QQQ].iloc[-2]]
        # reset portfolio value and qqq benchmark annually
        if self.Time.year!=self.year: self.benchmarks = [self.benchmarks[0], self.Portfolio.TotalPortfolioValue, self.history[self.QQQ].iloc[-2]]
        self.year = self.Time.year
        
        # SPY benchmark for main chart
        spy_perf = self.history[self.MRKT].iloc[-1] / self.benchmarks[0] * self.cap
        self.Plot('Strategy Equity', 'SPY', spy_perf)
        
        # Leverage gauge: cash level
        self.Plot('Cash level', 'cash', round(self.Portfolio.Cash+self.Portfolio.UnsettledCash, 0))
        
        # Annual saw tooth return comparison: Portfolio VS QQQ
        saw_portfolio_return = self.Portfolio.TotalPortfolioValue / self.benchmarks[1] - 1
        saw_qqq_return = self.history[self.QQQ].iloc[-1] / self.benchmarks[2] - 1
        self.Plot('Annual Saw Tooth Returns: Portfolio VS QQQ', 'Annual portfolio return', round(saw_portfolio_return, 4))
        self.Plot('Annual Saw Tooth Returns: Portfolio VS QQQ', 'Annual QQQ return', round(float(saw_qqq_return), 4))
        
        ### IN/Out indicator and wait days
        self.Plot("In Out", "in_market", int(self.be_in))
        self.Plot("Wait Days", "waitdays", self.adjwaitdays)
   
        
class SymbolData(object):
    def __init__(self, symbol, roc, algorithm):
        self.Symbol = symbol
        self.Roc = RateOfChange(roc)
        self.algorithm = algorithm
        
        self.consolidator = algorithm.ResolveConsolidator(symbol, Resolution.Daily)
        algorithm.RegisterIndicator(symbol, self.Roc, self.consolidator)
        
    def Warmup(self, history):
        for index, row in history.iterrows():
            self.Roc.Update(index, row['close'])