The strategy invests in 27 commodities, sorting them monthly by spreading pressure. It goes long on the three with the lowest pressure and short on the three with the highest, rebalancing monthly.
ASSET CLASS: CFDs, futures | REGION: Global | FREQUENCY:
Monthly | MARKET: commodities | KEYWORD: Commodity, Futures
I. STRATEGY IN A NUTSHELL
Invests in 27 commodities across Energy, Grains, Meats, Metal, and Softs. Spreading pressure—speculator spread positions divided by open interest—is averaged over 12 months. Each month, the strategy goes long the three commodities with the lowest spreading pressure and short the three with the highest. Positions are equally weighted and rebalanced monthly to exploit pricing inefficiencies linked to spread positioning.
II. ECONOMIC RATIONALE
Spreading pressure captures speculative hedging activity and reflects expectations about futures price dynamics, liquidity, and term structure. Rising spreading pressure since 2005 has reduced returns from carry and momentum strategies as arbitrage capital expanded, compressing anomalies. High spreading pressure thus signals diminished profitability, while low spreading pressure identifies commodities with superior return potential.
III. SOURCE PAPER
Speculator Spreading Pressure and the Commodity Futures Risk Premium [Click to Open PDF]
Yujing Gong, Arie E. Gozluklub, Gi H. Kim
<Abstract>
This paper investigates the impact of speculators’ trading activities on the commodity futures risk premium. In particular, we focus on speculators’ spread positions, and study the asset pricing implications of spreading pressure on the cross-section of commodity futures returns. We document that spreading pressure negatively predicts futures excess returns even after controlling for well-known determinants of futures returns such as basis-momentum. Furthermore, the spreading pressure factor-mimicking portfolio carries a significant risk premium of 21.55% per annum after commodity market financialization. Our single-factor model provides a better cross-sectional fit than the existing 2-factor or 3-factor models in the literature. We interpret these results as spreading pressure reflecting speculators’ expectation on the change in the slope and curvature of futures term structures and our spreading pressure factor linking to innovations in real economic uncertainty.
IV. BACKTEST PERFORMANCE
| Annualised Return | 21.19% |
| Volatility | 24.42% |
| Beta | 0.08 |
| Sharpe Ratio | 0.87 |
| Sortino Ratio | -0.21 |
| Maximum Drawdown | N/A |
| Win Rate | 51% |
V. FULL PYTHON CODE
import numpy as np
from AlgorithmImports import *
import data_tools
class SpeculatorSpreadingPressureandtheCommodityFuturesRiskPremium(QCAlgorithm):
def Initialize(self):
self.SetStartDate(2000, 1, 1)
self.SetCash(100000)
self.symbols = [
"CME_S1", # Soybean Futures, Continuous Contract
"CME_W1", # Wheat Futures, Continuous Contract
"CME_SM1", # Soybean Meal Futures, Continuous Contract
"CME_BO1", # Soybean Oil Futures, Continuous Contract
"CME_C1", # Corn Futures, Continuous Contract
"CME_O1", # Oats Futures, Continuous Contract
"CME_LC1", # Live Cattle Futures, Continuous Contract
"CME_FC1", # Feeder Cattle Futures, Continuous Contract
"CME_LN1", # Lean Hog Futures, Continuous Contract
"CME_GC1", # Gold Futures, Continuous Contract
"CME_SI1", # Silver Futures, Continuous Contract
"CME_PL1", # Platinum Futures, Continuous Contract
"CME_CL1", # Crude Oil Futures, Continuous Contract
"CME_HG1", # Copper Futures, Continuous Contract
"CME_LB1", # Random Length Lumber Futures, Continuous Contract
# "CME_NG1", # Natural Gas (Henry Hub) Physical Futures, Continuous Contract
"CME_PA1", # Palladium Futures, Continuous Contract
"CME_RR1", # Rough Rice Futures, Continuous Contract
"CME_DA1", # Class III Milk Futures
"ICE_RS1", # Canola Futures, Continuous Contract
"ICE_GO1", # Gas Oil Futures, Continuous Contract
"CME_RB2", # Gasoline Futures, Continuous Contract
"CME_KW2", # Wheat Kansas, Continuous Contract
"ICE_WT1", # WTI Crude Futures, Continuous Contract
"ICE_CC1", # Cocoa Futures, Continuous Contract
"ICE_CT1", # Cotton No. 2 Futures, Continuous Contract
"ICE_KC1", # Coffee C Futures, Continuous Contract
"ICE_O1", # Heating Oil Futures, Continuous Contract
"ICE_OJ1", # Orange Juice Futures, Continuous Contract
"ICE_SB1" # Sugar No. 11 Futures, Continuous Contract
]
self.period = 12 * 4
self.traded_count = 3
# Spreading pressure weekly data.
self.spreading_pressure = {}
for symbol in self.symbols:
# Add quantpedia back-adjusted data.
data = self.AddData(data_tools.QuantpediaFutures, symbol, Resolution.Daily)
data.SetFeeModel(data_tools.CustomFeeModel())
data.SetLeverage(5)
# COT data import.
cot_symbol = 'Q' + symbol.split('_')[1][:-1]
self.AddData(data_tools.CommitmentsOfTraders, cot_symbol, Resolution.Daily)
self.spreading_pressure[symbol] = RollingWindow[float](self.period)
self.recent_month = -1
def OnData(self, data):
for symbol in self.symbols:
cot_symbol = 'Q' + symbol.split('_')[1][:-1]
# Check if custom data is still coming.
if any([self.securities[x].get_last_data() and self.time.date() > data_tools.LastDateHandler.get_last_update_date()[x] for x in [symbol, cot_symbol]]):
self.liquidate(symbol)
self.spreading_pressure[symbol].reset()
continue
if cot_symbol in data and data[cot_symbol]:
cot_data = self.Securities[cot_symbol].GetLastData()
if cot_data:
speculator_long_count = cot_data.GetProperty("LARGE_SPECULATOR_LONG")
speculator_short_count = cot_data.GetProperty("LARGE_SPECULATOR_SHORT")
hedger_long_count = cot_data.GetProperty("COMMERCIAL_HEDGER_LONG")
hedger_short_count = cot_data.GetProperty("COMMERCIAL_HEDGER_SHORT")
trader_long_count = cot_data.GetProperty("SMALL_TRADER_LONG")
trader_short_count = cot_data.GetProperty("SMALL_TRADER_SHORT")
spread = min(speculator_long_count, speculator_short_count)
open_interest = speculator_long_count + speculator_short_count + hedger_long_count + hedger_short_count + trader_long_count + trader_short_count
if spread != 0 and open_interest != 0:
self.spreading_pressure[symbol].Add(spread / open_interest)
if self.recent_month == self.Time.month:
return
self.recent_month = self.Time.month
sorted_by_avg_spreading_pressure = sorted([x for x in self.spreading_pressure.items() if x[1].IsReady], key = lambda y: np.mean([x for x in y[1]]), reverse = True)
long = []
short = []
if len(sorted_by_avg_spreading_pressure) >= self.traded_count*2:
long = [x[0] for x in sorted_by_avg_spreading_pressure[-self.traded_count:]]
short = [x[0] for x in sorted_by_avg_spreading_pressure[:self.traded_count]]
# Trade execution.
targets: List[PortfolioTarget] = []
for i, portfolio in enumerate([long, short]):
for symbol in portfolio:
if symbol in data and data[symbol]:
targets.append(PortfolioTarget(symbol, ((-1) ** i) / len(portfolio)))
self.SetHoldings(targets, True)