Testing Indicator Sequences Against Subsequent Price Extremes

Summary

The document outlines an exploratory method for studying whether historical indicator patterns relate to later price movement. It encodes a sequence of MACD values over a lookback window, uses that sequence to name a CSV file, and records how often the pattern occurs along with the subsequent upward and downward price extremes over a forward interval. The approach is presented as an initial attempt to build a grid-based expert advisor from recurring signal sequences.

The author says the resulting examples did not produce compelling performance, with profitability rarely exceeding a stated ratio of 1.1, and invites others to improve the approach. This is a brief code-oriented sketch rather than a complete research protocol: it does not specify costs, out-of-sample evaluation, parameter selection safeguards, or how patterns are matched robustly. Its main lesson is that historical pattern counts and forward extrema alone do not demonstrate a tradable edge.

Key ideas

• The method encodes a window of MACD readings as a pattern identifier.
• For each observed pattern, it records occurrence counts and later upward and downward price extremes.
• Occurrence counts are intended to indicate the statistical weight of each pattern.
• The author reports weak example profitability and gives no evidence of robust out-of-sample performance.

Tags

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