Diversified Time-Series Momentum with Inverse-Volatility Sizing

Summary

This strategy takes monthly long or short positions across a diversified set of commodity, currency, equity-index, and government-bond futures. For each contract, it compares its excess return over the previous 12 months with zero, going long after a positive return and short after a negative one. Position weights are inversely related to volatility, and the long and short groups are scaled separately toward a stated annual volatility target, subject to a leverage cap.

The accompanying implementation uses historical volatility in place of the GARCH estimate described in its source discussion, and rebalances monthly. It supplies code, an instrument universe, data handling for continuous futures, and a fee model, but reports no backtest performance or comparative evidence. Results would depend on the historical sample, futures data construction, volatility estimates, trading costs, and leverage settings; the code is an implementation example rather than evidence that the strategy will work prospectively.

Key ideas

• The strategy takes long or short futures positions according to each contract’s trailing 12-month excess return.
• It allocates inversely to estimated volatility and scales long and short books separately toward a volatility target.
• The implementation substitutes historical volatility for the GARCH estimate mentioned in the strategy description.
• The code specifies monthly rebalancing, a broad multi-asset futures universe, and a leverage cap, but gives no performance results.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.