Discussing the article: "MQL5 Wizard Techniques you should know (Part 74): Using Patterns of Ichimoku and the ADX-Wilder with Supervised Learning"

[MetaQuotes]

Check out the new article: MQL5 Wizard Techniques you should know (Part 74): Using Patterns of Ichimoku and the ADX-Wilder with Supervised Learning.

We follow up on our last article, where we introduced the indicator pair of the Ichimoku and the ADX, by looking at how this duo could be improved with Supervised Learning. Ichimoku and ADX are a support/resistance plus trend complimentary pairing. Our supervised learning approach uses a neural network that engages the Deep Spectral Mixture Kernel to fine tune the forecasts of this indicator pairing. As per usual, this is done in a custom signal class file that works with the MQL5 wizard to assemble an Expert Advisor.

In the last article, we looked at the indicator pairing of Ichimoku and the ADX-Wilder as a complimentary S/R and trend tool pairing. As usual, we tested these in a wizard assembled Expert Advisor and looked at 10 different signal patterns. For this indicator pairing, most of them were able to forward walk profitably for 1-year, having done testing/ optimization on the previous year. However, there were 3 that did not do so satisfactorily and these were pattern-0, pattern-1, and pattern-5. We therefore follow up that article by examining if supervised learning can make a difference to their performance. Our approach in this is to reconstruct the signals of each of these patterns as a simple input vector into a neural network, essentially making the neural network an extra filter to the signal.

Author: Stephen Njuki