Discussing the article: "MQL5 Wizard Techniques you should know (Part 08): Perceptrons"
What is the time frame used in this test?
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 08): Perceptrons.
Perceptrons, single hidden layer networks, can be a good segue for anyone familiar with basic automated trading and is looking to dip into neural networks. We take a step by step look at how this could be realized in a signal class assembly that is part of the MQL5 Wizard classes for expert advisors.
The MQL5 wizard Expert-Signal class comes with a lot of example instances under the folder “Include\Expert\Signal” and each one of them can be used independently or combined with each other in putting together an Expert Advisor in the Wizard. For this article we will aim to create and use one such file in an expert adviser. This approach besides minimizing preliminary coding efforts, it allows testing more than one signal in a single expert advisor by attributing weighting to each used signal.
The Alglib perceptron classes are presented in extensive and interlinked network interfaces within the file “Include\Math\Alglib\dataanalysis.mqh”. It is easy to be overwhelmed when you first take a look, but we’ll look at a few critical classes here that hopefully will make this area easy to navigate.
The main motivation for using these Alglib classes to develop an Expert Advisor is the same for using the MQL5 wizard which is, idea testing. How can I succinctly determine if an idea x, or an input data set y is worth my effort in seriously developing further into a trading system? What we explore here could help in answering this question.
Before we jump in though it may be helpful to make a broader case on why perceptrons and perhaps neural networks in general are gaining a lot of traction in many circles. If we stick to finance and the markets we can see there are quite a few challenges in forecasting market action and the limitations of traditional analysis are arguably implicit in this.
Author: Stephen Njuki