Time Series Momentum Across Futures Markets
Summary
The document explains time series momentum as a strategy that uses each instrument’s own past return, rather than ranking assets against one another. Its central signal is the sign of the prior 12-month excess return: go long when positive and short when negative. The example universe spans commodity, currency, equity-index, and government-bond futures. Positions are scaled inversely to estimated volatility and rebalanced monthly; the source study used a univariate GARCH model, while the page notes historical volatility as a simpler alternative.
The cited research reports return persistence over intermediate horizons and partial reversal over longer ones, and describes diversified portfolios as having low correlation with passive benchmarks and favorable performance in extreme markets. The proposed behavioral explanation is initial underreaction followed by delayed overreaction. These are summaries of cited studies, not guarantees. The page also flags sensitivity to signal construction, volatility estimation, and even the starting day in some implementations, so results may vary with design choices and sample period.
Key ideas
- The strategy takes long or short positions based on the sign of each asset’s trailing 12-month excess return.
- The example applies the signal monthly to a diversified set of futures markets.
- Position sizes are inversely related to estimated volatility, with GARCH used in the source study.
- Research cited in the page reports intermediate-term persistence and partial reversal at longer horizons.
- The cited evidence suggests diversification and possible help during extreme equity markets, while design choices can affect results.
Tags
Cited by
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.