The strategy uses futures across commodities, equity indices, fixed income, and foreign exchange. It calculates expected returns based on historical data and applies a long-short strategy, rebalanced monthly with equally weighted positions.
ASSET CLASS: CFDs, futures | REGION: Global | FREQUENCY: Monthly | MARKET: bonds, commodities, currencies, equities | KEYWORD: Basis Indicator
I. STRATEGY IN A NUTSHELL
This strategy trades 65 global futures across commodities, equity indices, bonds, and FX. It predicts returns using the basis (difference between spot and futures prices) and its historical patterns. Positions are taken long if predicted returns are above the mean and short if below, with monthly rebalancing.
II. ECONOMIC RATIONALE
The strategy leverages the predictive power of the basis for future returns. Pooling information across markets enhances accuracy, and while restricted regression improves precision, simple historical averaging (naive predictors) often yields nearly as strong and more practical forecasts.
III. SOURCE PAPER
Predicting Out-of-Sample Returns: Using Basis to Beat the Historical Average[Click to Open PDF]
Marat Molyboga.
<Abstract>
This paper introduces an adaptive predictor that pools information across securities in four major asset classes (commodities, equities, fixed income and foreign exchange) while imposing restrictions on the sign and magnitude of coefficients in return forecasts. I demonstrate that the basis between spot and futures contracts predicts future returns across the asset classes. The predictor consistently beats the historical average, producing a median monthly out-of-sample R2, measured over the period between January 1986 and December 2016, of approximately 0.36%, a value that is comparable to those of the best equity premium predictors considered in Campbell and Thompson (2008). A simple long-short strategy based on the new predictor delivers an out-of-sample alpha of 2.5%-4.5% per annum with respect to the asset pricing models considered and produces an out-of-sample Sharpe ratio of almost 0.5, which is particularly striking since the strategy is countercyclical. This performance is robust across sub-periods, market environments, portfolio construction methodologies and transaction costs. A cross-sectional structure analysis reveals that two observable common factors, constructed as equally-weighted indices of the bases of financial assets and commodities, are related to the short-term interest rate and the business cycle, respectively.
IV. BACKTEST PERFORMANCE
| Annualised Return | 3.78% |
| Volatility | 7.67% |
| Beta | N/A |
| Sharpe Ratio | 0.49 |
| Sortino Ratio | N/A |
| Maximum Drawdown | N/A |
| Win Rate | N/A |
