Ehlers Fisher Transform for Reversal-Based Long and Short Signals

Summary

This strategy applies the Fisher Transform to normalized midpoint prices over a lookback window. It smooths the normalized value, constrains it near its allowable bounds, then transforms it; the direction of change in the resulting indicator determines whether the strategy holds long or short. A reverse-trading option flips those directions. The documented default lookback is 10 periods.

The article explains the transform as a way to make extreme normalized readings stand out and treats indicator turning points as possible price reversals. It cautions that the transformed series may not follow a Gaussian distribution, especially during unusual moves such as gaps, and recommends parameter tuning, filters, and stop losses. A BTC/USDT futures backtest configuration is included, but no performance results are given. The source changes position based on whether the Fisher value rises or falls, so it may switch frequently; it does not include a stop-loss rule despite discussing one as a possible improvement. The claimed accuracy is therefore not substantiated by the supplied evidence.

Key ideas

• The Fisher Transform is applied to a smoothed, normalized midpoint price over a selected lookback.
• An increase or decrease in the transformed value sets the long or short direction, with an option to reverse it.
• Unusual price moves can produce misleading signals, and the transformed data need not be Gaussian.
• The source contains no stop-loss logic and the included backtest setup reports no performance results.

Tags

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