Bitwise Perceptron Signals from 64-Bar Price Patterns

Summary

The article describes an MQL5 entry signal that encodes each of the last 64 bars as a binary value: bullish bars set a bit, while other bars leave it unset. A perceptron then adds the weights for active bits to a bias to produce an activation score. The proposed design aims to reduce the data and computation used in a tick-by-tick entry check, though it still iterates over bars and bits; claims of near-zero latency are not established by the method alone.

A second gate uses a pattern map to screen setups before the perceptron applies a threshold to decide whether to trade. The document reports forward tests in which the hybrid traded less often and had a smoother equity curve, with drawdown falling from 18.66% to 16.75%, while the algorithm-only version had slightly higher gross profit. These results illustrate a selectivity trade-off, not proof of general effectiveness. The binary encoding also loses price detail and treats flat bars like bearish ones, and the article calls for measuring latency and out-of-sample performance.

Key ideas

• The method maps bullish or non-bullish outcomes across 64 bars into a 64-bit state.
• A perceptron scores that state by summing weights for active bits with a bias.
• An algorithmic pattern-map gate screens states before the perceptron threshold is applied.
• The reported hybrid forward test reduced drawdown but also traded less frequently.
• Binary bar encoding discards price magnitude and can misclassify flat bars as bearish.

Tags

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