Discussing the article: "MQL5 Wizard Techniques you should know (Part 72): Using Patterns of MACD and the OBV with Supervised Learning"

[MetaQuotes]

Check out the new article: MQL5 Wizard Techniques you should know (Part 72): Using Patterns of MACD and the OBV with Supervised Learning.

We follow up on our last article, where we introduced the indicator pair of the MACD and the OBV, by looking at how this pairing could be enhanced with Machine Learning. MACD and OBV are a trend and volume complimentary pairing. Our machine learning approach uses a convolution neural network that engages the Exponential kernel in sizing its kernels and channels, when fine-tuning the forecasts of this indicator pairing. As always, this is done in a custom signal class file that works with the MQL5 wizard to assemble an Expert Advisor.

n our last article, where we introduced the indicator pair of the MACD and On-Balance-Volume oscillator (OBV) we examined 10 possible signals that could be generated from the complimentary pairing of these indicators. We always test 10 signal patterns as derived from a 2-indicator-pairing by performing a forward walk test for 1-year, having done training or optimization over the previous year. With the last article, only one signal pattern, pattern-7, was able to forward walk profitably. We went over some of the reasons why this indicator pairing performed dismally when compared to other indicators we have covered, however this also presented an ‘opportunity’ for us to explore machine learning in enhancing some of these signal patterns.

We therefore follow up that article by examining how, if at all, a CNN network that uses the rational quadratic kernel, could help and perhaps improve indicator signal interpretation and application. The Rational Quadratic (RQ) kernel is defined by the following formula:

Author: Stephen Njuki