Discussing the article: "MQL5 Wizard Techniques you should know (Part 85): Using Patterns of Stochastic-Oscillator and the FrAMA with Beta VAE Inference Learning"
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 85): Using Patterns of Stochastic-Oscillator and the FrAMA with Beta VAE Inference Learning.
This piece follows up ‘Part-84’, where we introduced the pairing of Stochastic and the Fractal Adaptive Moving Average. We now shift focus to Inference Learning, where we look to see if laggard patterns in the last article could have their fortunes turned around. The Stochastic and FrAMA are a momentum-trend complimentary pairing. For our inference learning, we are revisiting the Beta algorithm of a Variational Auto Encoder. We also, as always, do the implementation of a custom signal class designed for integration with the MQL5 Wizard.
We are using the indicator-pairing of our last installment, part-84, where we as usual considered 10 key, distinct patterns that we indexed from 0 to 9 having been derived from the Stochastic Oscillator and the Fractal Adaptive Moving Average. We noted varying performance across these 10 patterns in the walk-forward tests we made. Patterns from 0 to 4 as well as those from 7 to 8 showed some robustness by being profitable across our scope of assets that had been chosen to capitalize on different market regimes. Nonetheless, patterns 5, 6 and 9 lagged majorly, failing to show profitability in the out-of-sample forward walk period.
It is worth re-emphasizing that our test window is very limited, therefore these results at best should be taken as a clue on what patterns need further-testing, not which patterns are dependable. The underperformer patterns were defined as flat FrAMA with Stochastic crossovers for pattern-5, overbought/oversold Stochastic hooks with sloping FrAMA for pattern-6, and extreme Stochastic-Oscillator levels with opposing FrAMA slopes for pattern-9.
These failures could be attributed to limitations of the inflexibility of these patterns. Or it could be that testing scope was too restrictive, and it is these patterns and not some of our ‘profitable’ ones above that are resilient in the long term. This debate can best be settled by the reader from independent testing. For our purposes, though, we look into how machine learning could, if at all, rehabilitate the fortunes of patterns 5, 6, and 9.
Buy signal (Pattern 5): Flat FrAMA + Stoch cross up under 30
Author: Stephen Njuki