A Gaussian Z-Score Strategy with Adaptive Thresholds

Summary

This strategy applies a Z-score to a moving average of Heikin-Ashi closing prices, then smooths the score with another exponential average. It sets dynamic upper, lower, and middle levels from the score curve's recent distribution. Crossings of these levels, along with recent score extremes, define long and short signals. The document frames this combination as a way to identify unusual price behavior and possible reversals, while proposing walk-forward testing, alternative thresholds, and other price inputs as areas to investigate.

The published test configuration uses BTC-USDT futures on a five-minute chart with one-minute base data for about a week, but the document reports no performance statistics. The source itself describes the strategy as an attempt and notes that it may not behave as intended. The method is sensitive to its smoothing windows and threshold choices; Heikin-Ashi data can lag, and extreme-value signals may be unreliable. The stated benefits, including filtering false breaks, are hypotheses rather than demonstrated results, so broader testing is needed before drawing conclusions.

Key ideas

• The method calculates a Z-score from a smoothed Heikin-Ashi price series and smooths that score again.
• Recent score percentiles provide dynamic thresholds for crossover-based entries and exits.
• Recent score highs and lows are also used to identify potential reversal signals.
• The document flags parameter sensitivity, lag, and unreliable extreme-value signals as limitations.
• Its short BTC-USDT futures test reports no results, so the proposed benefits are not established.

Tags

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