Understanding Trend Following: A Quantitative Approach

Trend following is one of the most enduring systematic trading strategies in quantitative finance. At its core, it rests on a deceptively simple premise: markets trend. Prices, rather than moving randomly, exhibit persistent directional movements that can be captured and exploited.

The Mathematics of Momentum

The foundation of trend following lies in momentum—the tendency for assets that have performed well to continue performing well, and vice versa. We can express this mathematically through various signal constructions.

Simple Moving Average Crossover

The most intuitive trend signal uses moving average crossovers. Given a price series $P_t$ , we define the simple moving average as:

SMA_n(t) = \frac{1}{n} \sum_{i=0}^{n-1} P_{t-i}

A basic trend signal $S_t$ can be constructed as:

S_t = \text{sign}(SMA_{fast}(t) - SMA_{slow}(t))

Where $S_t = 1$ indicates a long position and $S_t = -1$ indicates a short position.

Exponential Moving Average

For more responsive signals, traders often prefer exponential moving averages, which weight recent observations more heavily:

EMA_t = \alpha \cdot P_t + (1 - \alpha) \cdot EMA_{t-1}

Where the smoothing factor $\alpha = \frac{2}{n+1}$ and $n$ is the effective lookback period.

Risk-Adjusted Returns

Raw momentum signals must be scaled by volatility to ensure consistent risk exposure. The Sharpe Ratio provides a measure of risk-adjusted performance:

SR = \frac{E[R_p - R_f]}{\sigma_p}

Where:

$R_p$ is the portfolio return
$R_f$ is the risk-free rate
$\sigma_p$ is the portfolio standard deviation

Most trend followers target annualized Sharpe ratios between 0.5 and 1.5, which requires careful position sizing.

Position Sizing with Volatility Targeting

To maintain consistent risk, positions are scaled inversely with volatility:

\text{Position Size} = \frac{\text{Target Volatility}}{\text{Asset Volatility}} \times \text{Signal Strength}

This approach, pioneered by managed futures funds, ensures that a highly volatile asset receives a smaller position than a stable one, all else equal.

The Time Series Momentum Factor

Academic research, notably by Moskowitz, Ooi, and Pedersen (2012), formalized trend following as time series momentum. The return predictor is simply the asset’s own past return:

r_{t,t+1} = \alpha + \beta \cdot \text{sign}(r_{t-12m,t}) + \epsilon_t

Their findings showed significant positive $\beta$ across asset classes—equities, bonds, commodities, and currencies—suggesting momentum is a pervasive phenomenon.

Why Do Trends Persist?

Several theories explain the persistence of trends:

Behavioral biases: Investors underreact to new information, causing prices to adjust slowly to fundamental values.
Institutional frictions: Large investors cannot rebalance instantly, creating predictable price pressure.
Central bank policies: Monetary policy operates with long and variable lags, inducing trending behavior in rates and currencies.
Herding: Market participants observe and follow each other, amplifying initial price movements.

Practical Considerations

Drawdowns and Patience

Trend following strategies experience significant drawdowns, often 20-40% peak-to-trough. The strategy’s edge lies in capturing the fat tails of return distributions—the occasional large trend that generates outsized profits.

Correlation Benefits

Trend following exhibits crisis alpha—the tendency to perform well during market dislocations. This makes it valuable as a portfolio diversifier, particularly for traditional stock/bond allocations.

Transaction Costs

Frequent signal changes generate transaction costs that erode returns. Sophisticated implementations use:

Signal smoothing to reduce turnover
Execution algorithms to minimize market impact
Position buffers to avoid trading on noise

Conclusion

Trend following represents a systematic approach to capturing persistent price movements. While the mathematics can become sophisticated—incorporating time-varying volatility, cross-sectional signals, and machine learning enhancements—the core insight remains: the trend is your friend.

For quantitative practitioners, the challenge lies not in identifying trends, but in managing risk, controlling costs, and maintaining discipline through inevitable drawdown periods.

Next in this series: We’ll examine cross-sectional momentum strategies and their interaction with value factors.