Equity · United States · CFDs, ETFs, futures · Daily · United States Free

US Holiday Effect in EU Markets

Academic paper

Paper linkhttp://www.efmaefm.org/0EFMAMEETINGS/EFMA%20ANNUAL%20MEETINGS/2011-Braga/papers/0465.pdf

Strategy in a nutshell

The strategy exploits informational advantages in the options market to forecast stock returns. By analyzing option-based indicators—such as volatility skew, implied-realized volatility spread, and changes in skew—it identifies stocks likely to outperform or underperform. The portfolio goes long on stocks with bullish option signals and short on those with bearish ones, rebalanced weekly for consistency.

Economic rationale

Options markets often reflect private information earlier than equities because of their lower transaction costs, higher leverage, and relaxed short-selling limits. When equity markets lag in incorporating this information, option-implied signals can predict future stock movements.

Backtest performance

Annualised return6.52%

Beta0

Sortino ratio-0.451

Win rate59%

Full Python code

1from AlgorithmImports import *
2class USHolidayEUMarkets(QCAlgorithm):
3def Initialize(self):
4 self.SetStartDate(2012, 1, 1)
5 self.SetCash(100000)           
6 
7 data = self.AddData(QuantpediaFutures, "EUREX_FDAX1", Resolution.Daily)
8 data.SetFeeModel(CustomFeeModel())
9 self.symbol = data.Symbol
10 
11def OnData(self, data):
12 if self.securities[self.symbol].get_last_data() and self.time.date() > QuantpediaFutures.get_last_update_date()[self.symbol]:
13     self.liquidate()
14     return
15 calendar1 = self.TradingCalendar.GetDaysByType(TradingDayType.PublicHoliday, self.Time, self.Time)
16 calendar2 = self.TradingCalendar.GetDaysByType(TradingDayType.Weekend, self.Time, self.Time + timedelta(days=2))
17 
18 holidays = [i.Date for i in calendar1]
19 weekends = [i.Date for i in calendar2]
20 
21 # subtract weekends in all holidays
22 public_holidays = list(set(holidays) - set(weekends))
23 if data.contains_key(self.symbol) and data[self.symbol]:
24     if not self.Portfolio.Invested and len(public_holidays) > 0:
25         self.SetHoldings(self.symbol, 1)
26     if self.Portfolio.Invested and len(public_holidays) == 0:
27         self.Liquidate()
28# Quantpedia data.
29# NOTE: IMPORTANT: Data order must be ascending (datewise)
30class QuantpediaFutures(PythonData):
31_last_update_date:Dict[Symbol, datetime.date] = {}
32@staticmethod
33def get_last_update_date() -> Dict[Symbol, datetime.date]:
34return QuantpediaFutures._last_update_date
35def GetSource(self, config, date, isLiveMode):
36 return SubscriptionDataSource("data.quantpedia.com/backtesting_data/futures/{0}.csv".format(config.Symbol.Value), SubscriptionTransportMedium.RemoteFile, FileFormat.Csv)
37def Reader(self, config, line, date, isLiveMode):
38 data = QuantpediaFutures()
39 data.Symbol = config.Symbol
40 
41 if not line[0].isdigit(): return None
42 split = line.split(';')
43 
44 data.Time = datetime.strptime(split[0], "%d.%m.%Y") + timedelta(days=1)
45 data['back_adjusted'] = float(split[1])
46 data['spliced'] = float(split[2])
47 data.Value = float(split[1])
48 if config.Symbol not in QuantpediaFutures._last_update_date:
49     QuantpediaFutures._last_update_date[config.Symbol] = datetime(1,1,1).date()
50 if data.Time.date() > QuantpediaFutures._last_update_date[config.Symbol]:
51     QuantpediaFutures._last_update_date[config.Symbol] = data.Time.date()
52 return data
53# Custom fee model.
54class CustomFeeModel(FeeModel):
55def GetOrderFee(self, parameters):
56 fee = parameters.Security.Price * parameters.Order.AbsoluteQuantity * 0.00005
57 return OrderFee(CashAmount(fee, "USD"))