Discussing the article: "MQL5 Wizard Techniques you should know (Part 70): Using Patterns of SAR and the RVI with a Exponential Kernel Network"
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Check out the new article: MQL5 Wizard Techniques you should know (Part 70): Using Patterns of SAR and the RVI with a Exponential Kernel Network.
We follow up our last article, where we introduced the indicator pair of the SAR and the RVI, by considering how this indicator pairing could be extended with Machine Learning. SAR and RVI are a trend and momentum 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.
In the last article, we introduced this complimentary pair of the Parabolic SAR indicator (SAR) and the Relative Vigour Index oscillator (RVI). From our testing of 10 patterns, three failed to perform clean forward walks; namely those indexed 1, 2, and 6. Our indexing of these patterns from 0 to 9 allows us to easily compute the map value that allows their exclusive use by the Expert Advisor. For instance, if a pattern is indexed 1 then we have to set the parameter ‘PatternsUsed’ to 2 to the power 1 which comes to 2.
If the index is 2 then this is 2 to the power 2 which comes to 4, and so on. The maximum value that this parameter can be assigned, meaningfully, is 1023 since we have only 10 parameters. Any number between 0 and 1023 that is not a pure exponent of 2 would represent a combination of these patterns, and the reader could explore setting up the expert Advisor to use multiple patterns. However, based on our arguments and test results presented in past articles we choose not to explore this avenue within these series, for now.
As promised in one of the past recent articles, we are now going to attempt to resuscitate the three pattern signals 1, 2, and 6 that were not able to perform clean forward walks in the past article, with supervised learning. In applying machine learning to these MQL5 indicator signals, we resort to Python to help with coding and training a network model. This is because of the efficiencies it is able to provide, even without a GPU. When using Python, we rely on MetaTrader’s Python Module that allows us to connect to a MetaTrader Broker’s server once we provide a login username and password.
Author: Stephen Njuki