Momentum & Trend-Following Systems: A Comprehensive Guide

The Momentum Effect

The momentum effect — the tendency for assets that have recently performed well to continue performing well, and vice versa — is one of the most robust and widely documented anomalies in financial markets. First formally documented by Jegadeesh and Titman (1993), momentum has been observed across virtually every asset class, geographic region, and time period studied.

For educational purposes, it is important to note that while momentum has been historically persistent, it is subject to significant drawdowns and periodic reversals. The momentum factor experienced severe crashes in 2009 and other stress periods, highlighting that no factor provides guaranteed returns.

Types of Momentum

Cross-Sectional Momentum

Cross-sectional momentum ranks all securities in a universe by their recent returns (typically 2-12 months) and takes long positions in top performers and short positions in bottom performers. The most common implementation uses 12-month returns, excluding the most recent month to avoid short-term reversal effects.

Time-Series Momentum

Also called "trend following," time-series momentum compares each security's recent return to zero (or another threshold) rather than to other securities. If a security's recent return is positive, go long; if negative, go short or stay flat. This approach naturally adjusts exposure based on market conditions.

Dual Momentum

Popularized by Gary Antonacci, dual momentum combines both cross-sectional and time-series momentum. First, it applies a time-series filter (is absolute momentum positive?). Then, among assets with positive momentum, it selects those with the strongest relative momentum. This approach historically reduced drawdowns compared to pure momentum strategies.

Momentum Factor Construction

For educational research purposes, momentum factors are typically constructed as follows:

1. Formation period: Calculate returns over the lookback period (commonly 12 months minus the most recent month, known as "12-1" momentum).
2. Ranking: Rank all securities by their formation-period returns.
3. Portfolio construction: Form portfolios (quintiles or deciles) based on rankings.
4. Holding period: Hold positions for a specified period (commonly 1-3 months) before rebalancing.
5. Rebalancing: At the end of each holding period, repeat the ranking and portfolio formation process.

Trend-Following Implementation

Practical trend-following systems often use one or more of these approaches:

• Moving average rules: Go long when price is above its moving average, flat or short when below.
• Channel breakouts: Enter positions when price exceeds the highest high (or lowest low) of the past N periods.
• Momentum scoring: Combine multiple lookback periods into a composite momentum score.

Position Sizing

Many trend-following systems use volatility-based position sizing (inverse volatility weighting), which allocates more capital to lower-volatility positions and less to higher-volatility positions. This creates a more balanced risk contribution across positions.

Historical Characteristics

Academic research and educational analysis have identified several characteristics of momentum strategies in historical data:

• Persistent but cyclical: Momentum returns are positive on average across long periods but experience sharp drawdowns (momentum crashes).
• Negatively skewed: Momentum strategies tend to have frequent small wins and occasional large losses, creating negative skewness.
• Correlation with market regime: Momentum tends to perform poorly during sharp market reversals, particularly V-shaped recoveries.
• Diversification benefits: Momentum factors are relatively uncorrelated with traditional value factors, offering potential portfolio diversification.

Disclaimer: This guide is provided for educational purposes only. Momentum and trend-following strategies involve significant risks including the possibility of substantial losses. Historical patterns are not indicative of future results.