Multi-Timeframe Momentum Architecture: 12-Minus-1 Month vs Short-Term Trend Reversals

The Dual Dynamics of Price Persistence and Short-Term Reversal

Cross-sectional momentum is among the most extensively documented phenomena in empirical finance. Jegadeesh and Titman (1993) demonstrated that equities displaying superior performance over intermediate lookbacks (3 to 12 months) continue to outperform over subsequent holding periods.

However, when measuring equity price trajectories, quantitative researchers encounter a sharp structural paradox: at the 1-month horizon, equities exhibit a pronounced short-term reversal anomaly rather than trend continuation. A naive 12-month momentum calculation that includes the most recent month conflates two opposing market dynamics.

Microstructure Origins of the 1-Month Reversal

The 1-month reversal anomaly (Jegadeesh 1990; Lehmann 1990) is primarily driven by market microstructure frictions and short-term liquidity provision:

• Bid-Ask Bounce: End-of-month trade executions alternating between bid and ask prices create mechanical negative serial correlation in observed returns.
• Market Maker Inventory Balancing: Liquidity providers demand price concessions to absorb temporary order flow imbalances, causing temporary price dislocations that mean-revert over the subsequent weeks.
• Tax-Loss and Calendar Rebalancing: Systematic portfolio rebalancing and seasonal tax-loss selling create short-term selling pressure unrelated to fundamental economic information.

A standard $t-12$ to $t$ calculation allows this high-frequency liquidity noise to contaminate the intermediate momentum signal.

The 12-Minus-1 Architecture

To eliminate microstructure contamination, systematic momentum architectures employ the classic 12-minus-1 month lookback ($t-12$ to $t-1$):

$\text{Mom}_{12-1, t} = \prod_{\tau=2}^{12} (1 + R_{t-\tau}) - 1$

By omitting the immediate trailing month ($t-1$), quantitative factor models purge transient liquidity shocks and bid-ask bounce while preserving the underlying behavioral and underreaction drivers that power multi-month trend persistence.

Systematic portfolio managers review quantitative momentum specifications in the Parallax Momentum Factor Documentation and evaluate multi-timeframe factor exposures across portfolios using Parallax Features.