The strategy trades Chinese A-shares, using oil-driven momentum and volatility sorting, buying high-momentum, low-volatility stocks, selling low-momentum, high-volatility stocks, and adjusting positions monthly based on market conditions.
ASSET CLASS: stocks | REGION: China| FREQUENCY: Monthly | MARKET: equities | KEYWORD: China, Volatility, Momentum
I. STRATEGY IN A NUTSHELL
Trade A-shares using industry and stock-level momentum and volatility. Monthly, go long on high-momentum, low-volatility stocks and short low-momentum, high-volatility stocks, adjusting positions based on oil volatility.
II. ECONOMIC RATIONALE
Oil price volatility drives industry and stock momentum in China. Conditioning strategies on oil volatility and combining momentum with low-risk filters significantly enhances returns, alpha, and risk-adjusted performance.
III. SOURCE PAPER
Oil and Stock Market Momentum [Click to Open PDF]
Chun-Da Chen, Lamar University; Chiao-Ming Cheng, ZhiDao Investment Management; Riza Demirer, Southern Illinois University Edwardsville – Department of Economics & Finance; Economic Research Forum (ERF)
<Abstract>
This study provides a novel perspective to the oil-stock market nexus by examining the predictive ability of oil return and volatility on stock market momentum in China. We find that oil return volatility serves as a strong predictor of industry momentum, even after controlling for stock market state, volatility and key macroeconomic variables. We argue that the predictive ability of oil over momentum payoffs is driven by time-varying investor sentiment that relates to excess buying pressure on winner stocks during uncertain times, captured by oil return volatility. Our tests also show that an oil-based momentum strategy wherein the investor conditions the trade on the state of oil return volatility yields significant abnormal returns, more than double that could be obtained from the conventional momentum strategy. In short, the findings suggest that oil market dynamics can contribute to stock market inefficiencies in such a way that these inefficiencies create significant abnormal profits for active managers.
IV. BACKTEST PERFORMANCE
| Annualised Return | 62.88% |
| Volatility | 38.23% |
| Beta | 0.001 |
| Sharpe Ratio | 1.64 |
| Sortino Ratio | N/A |
| Maximum Drawdown | N/A |
| Win Rate | 48% |
V. FULL PYTHON CODE
from AlgorithmImports import *
import numpy as np
#endregion
class OilVolatilityAffectsIndustryMomentumInChina(QCAlgorithm):
def Initialize(self):
self.SetStartDate(2014, 1, 1) # China indexes starts since 2012
self.SetCash(100000)
self.tickers = [
'SZAI', 'SZBSI',
'SZC50', 'SZCCIP',
'SZCDSI', 'SZCI',
'SZCOI', 'SZCPI',
'SZCSI', 'SZCSSI',
'SZD50', 'SZDII',
'SZEMI', 'SZEPI',
'SZESI', 'SZEXT50',
'SZFSI', 'SZHC50',
'SZHCI', 'SZHCSI',
'SZIBIP', 'SZICI',
'SZIFI', 'SZISI',
'SZITI', 'SZMAI',
'SZMEI', 'SZMII',
'SZMNI', 'SZMSI',
'SZPI', 'SZRDI',
'SZREI', 'SZSRIP',
'SZTRI', 'SZTSI',
'SZTSSI', 'SZUII',
'SZUSI', 'SZWHI'
]
self.data = {}
self.symbols = []
self.period = 12 * 21 # 12 months of daily closes
self.oil_period = 36 * 21 # 3 years of daily closes
self.SetWarmUp(2*self.oil_period)
self.months_skip = 1 # Calculating performance of period from month t-12 to month t-2.
for ticker in self.tickers:
security = self.AddData(QuantpediaIndexChina, ticker, Resolution.Daily)
security.SetFeeModel(CustomFeeModel())
security.SetLeverage(5)
symbol = security.Symbol
self.symbols.append(symbol)
self.data[symbol] = SymbolData(self.period)
self.symbol = self.AddEquity('SPY', Resolution.Daily).Symbol
self.oil_symbol = self.AddData(QuantpediaFutures, 'CME_CL1', Resolution.Daily).Symbol
self.data[self.oil_symbol] = SymbolData(self.oil_period)
self.max_missing_days:int = 5
self.selection_flag = False
self.Schedule.On(self.DateRules.MonthStart(self.symbol), self.TimeRules.AfterMarketOpen(self.symbol), self.Selection)
def OnData(self, data):
for symbol in self.symbols + [self.oil_symbol]:
if symbol in data:
if data[symbol]:
self.data[symbol].update(data[symbol].Value)
if self.IsWarmingUp: return
if not self.selection_flag:
return
self.selection_flag = False
# oil data is still comming in
if self.Securities[self.oil_symbol].GetLastData() and (self.Time.date() - self.Securities[self.oil_symbol].GetLastData().Time.date()).days > self.max_missing_days:
self.Liquidate()
return
if self.data[self.oil_symbol].is_ready():
short_term_volatility = self.data[self.oil_symbol].volatility(self.period)
long_term_volatility = self.data[self.oil_symbol].volatility(self.oil_period)
performances = {}
for symbol in self.symbols:
# futures data is still comming in
if self.Securities[symbol].GetLastData() and (self.Time.date() - self.Securities[symbol].GetLastData().Time.date()).days <= self.max_missing_days:
if self.data[symbol].is_ready():
performances[symbol] = self.data[symbol].performance(self.months_skip)
performances_median = np.median([x[1] for x in performances.items()])
winners = [x[0] for x in performances.items() if performances[x[0]] >= performances_median]
losers = [x[0] for x in performances.items() if performances[x[0]] < performances_median]
if short_term_volatility > long_term_volatility: # High volatility
long = losers
short = winners
else: # Small volatility
long = winners
short = losers
# Trade Execution
stocks_invested = [x.Key for x in self.Portfolio if x.Value.Invested]
for symbol in stocks_invested:
if symbol not in long + short:
self.Liquidate(symbol)
long_length = len(long)
short_length = len(short)
for symbol in long:
if symbol in data and data[symbol]:
self.SetHoldings(symbol, 1 / long_length)
for symbol in short:
if symbol in data and data[symbol]:
self.SetHoldings(symbol, -1 / short_length)
def Selection(self):
self.selection_flag = True
class SymbolData():
def __init__(self, period):
self.closes = RollingWindow[float](period)
def update(self, close):
self.closes.Add(close)
def is_ready(self):
return self.closes.IsReady
def performance(self, months_skip):
closes = [x for x in self.closes]
closes = closes[months_skip * 21:] # 21 are total closes in one month
return (closes[0] - closes[-1]) / closes[-1]
def volatility(self, period):
closes = [x for x in self.closes]
monthly_closes = np.array(closes[:period])
monthly_returns = (monthly_closes[:-1] - monthly_closes[1:]) / monthly_closes[1:]
vol = np.std(monthly_returns)
return vol
# Custom fee model
class CustomFeeModel(FeeModel):
def GetOrderFee(self, parameters):
fee = parameters.Security.Price * parameters.Order.AbsoluteQuantity * 0.00005
return OrderFee(CashAmount(fee, "USD"))
# Quantpedia data.
# NOTE: IMPORTANT: Data order must be ascending (datewise)
class QuantpediaFutures(PythonData):
def GetSource(self, config, date, isLiveMode):
return SubscriptionDataSource("data.quantpedia.com/backtesting_data/futures/{0}.csv".format(config.Symbol.Value), SubscriptionTransportMedium.RemoteFile, FileFormat.Csv)
def Reader(self, config, line, date, isLiveMode):
data = QuantpediaFutures()
data.Symbol = config.Symbol
if not line[0].isdigit(): return None
split = line.split(';')
data.Time = datetime.strptime(split[0], "%d.%m.%Y") + timedelta(days=1)
data['back_adjusted'] = float(split[1])
data['spliced'] = float(split[2])
data.Value = float(split[1])
return data
# Quantpedia data
# NOTE: IMPORTANT: Data order must be ascending (datewise)
class QuantpediaIndexChina(PythonData):
def GetSource(self, config, date, isLiveMode):
source = "data.quantpedia.com/backtesting_data/index/china/{0}.csv".format(config.Symbol.Value)
return SubscriptionDataSource(source, SubscriptionTransportMedium.RemoteFile, FileFormat.Csv)
def Reader(self, config, line, date, isLiveMode):
data = QuantpediaIndexChina()
data.Symbol = config.Symbol
try:
if not line[0].isdigit(): return None
split = line.split(',')
data.Time = datetime.strptime(split[0], "%Y-%m-%d") + timedelta(days=1)
data['Price'] = float(split[1])
data.Value = float(split[1])
except:
return None
return data