Equity · Europe · CFDs, ETFs, futures · Weekly · Europe Free

Short-Term Reversal in Equity Index Futures

Academic paper

Paper linkhttp://centerforpbbefr.rutgers.edu/2006/Paper%202006/12AS01-046-INT_Index_Futures.pdf

Strategy in a nutshell

Trade seven EU index futures (Germany, France, Italy, Spain, Netherlands, Belgium, Austria) using nearest contracts. Classify countries as winners or losers based on the past five days’ returns. Go long on losers, short on winners, weighted by deviation from the benchmark. Rebalance weekly to capture mean-reversion.

Economic rationale

Short-term contrarian strategies exploit overreaction: past winners tend to underperform, and past losers tend to outperform. These predictable reversals provide opportunities to profit from temporary mispricing in EU index futures.

Backtest performance

Annualised return29.6%

Beta0.03

Sortino ratio-0.001

Win rate49%

Full Python code

1import numpy as np
2from AlgorithmImports import *
3class ShortTermReversal(QCAlgorithm):
4def Initialize(self):
5self.SetStartDate(2000, 1, 1)
6self.SetCash(100000)
7
8self.symbols = ['EWG', 'EWQ', 'EWI', 'EWP', 'EWN', 'EWK', 'EWO']
9
10# Daily price data.
11self.data = {}
12self.period = 5
13self.SetWarmUp(self.period)
14
15for symbol in self.symbols:
16    data = self.AddEquity(symbol, Resolution.Daily)
17    data.SetFeeModel(CustomFeeModel())
18    data.SetLeverage(5)
19    
20    self.data[symbol] = RollingWindow[float](self.period)
21
22self.Schedule.On(self.DateRules.Every(DayOfWeek.Thursday), self.TimeRules.AfterMarketOpen(self.symbols[0]), self.Rebalance)
23def OnData(self, data):
24for symbol in self.data:
25    symbol_obj = self.Symbol(symbol)
26    if symbol_obj in data.Keys:
27        if data[symbol_obj]:
28            price = data[symbol_obj].Value
29            if price != 0:
30                self.data[symbol].Add(price)
31def Rebalance(self):
32self.Liquidate()
33
34symbol_return = {}
35for symbol in self.symbols:
36    if self.data[symbol].IsReady: 
37        if self.Securities[symbol].GetLastData() and (self.Time.date() - self.Securities[symbol].GetLastData().Time.date()).days < 5:
38            closes = [x for x in self.data[symbol]]
39            symbol_return[symbol] = closes[0] / closes[-1] - 1
40
41if len(symbol_return) != 0:
42    avg_ret = np.average([x[1] for x in symbol_return.items()])
43    
44    # Average return weighting.
45    return_diff = {x[0] : abs(x[1] - avg_ret) for x in symbol_return.items()}
46
47    total_diff = sum(x[1] for x in return_diff.items())
48    weight_ratio = float(1 / total_diff)
49
50    # Trade execution.
51    winners = [x[0] for x in return_diff.items() if symbol_return[x[0]] > avg_ret]
52    losers = [x[0] for x in return_diff.items() if symbol_return[x[0]] < avg_ret]
53    
54    for symbol in winners + losers:
55        weight = return_diff[symbol] * weight_ratio
56        if symbol in winners:
57            self.SetHoldings(symbol, -weight)
58        elif symbol in losers:
59            self.SetHoldings(symbol, weight)
60
61# Custom fee model
62class CustomFeeModel(FeeModel):
63def GetOrderFee(self, parameters):
64fee = parameters.Security.Price * parameters.Order.AbsoluteQuantity * 0.00005
65return OrderFee(CashAmount(fee, "USD"))