Hull Moving Average Trend-Following Signals and Risks

Summary

This document describes a trend-following system that uses the Hull moving average (HMA) to generate long and short signals. The method compares weighted averages at half-length and full-length periods, smooths their difference, then trades when the resulting curve changes direction. Its example uses a period of 16, with an 8-period average and a 4-period smoothing step. Positions are sized as a share of equity, and the strategy can take either direction.

The discussion presents HMA as a quicker-moving alternative to ordinary moving averages, while noting that faster responses can also raise turnover and overtrading risk. It warns that trends can reverse abruptly, results may vary by instrument, and parameter tuning can overfit. Suggested safeguards include additional signal filters, stop losses, walk-forward parameter evaluation, and position sizing. The document supplies code and a short Bitcoin futures backtest configuration, but no performance results; its claimed advantages are therefore not demonstrated by reported evidence.

Key ideas

• The strategy uses a smoothed difference of half-length and full-length weighted moving averages to define the Hull curve.
• An upward or downward change in the curve triggers a long or short position.
• The example sets the main period to 16 and uses a 4-period smoothing step.
• The document identifies fast turnover, whipsaws, instrument variation, and parameter overfitting as risks.
• It proposes filters, stop losses, walk-forward analysis, and position sizing as possible safeguards.

Tags

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