Metatrader 5 / Metatrader 4 for MQL5 / MQL4 articles preview

Welcome to the second installment of our article series, "Modified Grid-Hedge EA in MQL5." Let's begin by recapping what we covered in the first part. In Part I, we explored the classic hedge strategy, automated it using an Expert Advisor (EA), and conducted tests in the strategy tester, analyzing some initial results. This marked the first step in our journey toward creating a Modified Grid-Hedge EA—a blend of classic hedge and grid strategies.

more...

We discussed in the previous article how to use socket (websocket) to communicate between EA and python server to solve the backtesting problem, and also discussed why we adopted this technique. In this article, we will discuss how to use onnx, which is natively supported by mql5, to perform inference with our model, but this method has some limitations. If your model uses operators that are not supported by onnx, it may end in failure, so this method is not suitable for all models (of course, you can also add operators to support your model, but it requires a lot of time and effort). This is why I spent a lot of space in the previous article to introduce the socket method and recommend it to you. Of course, converting a general model to onnx format is very convenient, and it provides us with effective support for cross-platform operations. This article mainly involves some basic operations of operating ONNX models in mql5, including how to match the input and output of torch models and ONNX models, and how to convert suitable data formats for ONNX models. Of course, it also includes EA order management. I will explain it in detail for you. Now let’s start the main topic of this article!

more...

Multi-currency Expert Advisor is an Expert Advisor or trading robot that can trade (open orders, close orders and manage orders, for example: Trailing Stop Loss and Trailing Profit) for more than 1 symbol pair from only one symbol chart, where in this article Expert Advisor will trade for 30 pairs. In this article we will use two RSI indicators with crossing signals, Fast RSI crossing with Slow RSI.

As has been proven in previous articles, we all know that multi-currency trading is already possible with the power, capabilities and facilities provided by MQL5, both in the trading terminal and in the strategy tester.

With the aim of meeting the important needs of traders who want an efficient and effective trading robot, so by relying on the power, capabilities and facilities provided by the very reliable MQL5, we can think of various ideas and strategies to create a simple multi-currency expert advisor, where in this article we will use Two RSI indicators cross each other's lines.

more...

Adaboost, short for adaptive boosting is an ensemble machine learning model that attempts to build a strong classifier out of weak classifiers.

more...

For a few time-series, it is possible to devise a formula for the next value in the sequences basing off of previous values that appeared within it. Number walls allow this to be accomplished by preliminarily generating a ‘wall of numbers’, in the form of a matrix via what is referred to as the cross-rule. In generating this matrix, the primary goal is to establish if the sequence in question is convergent and the number wall cross rule algorithm gladly answers this question, if after a few rows of application, the subsequent rows in the matrix are only zeroes.

more...

In this series, we have already examined a fairly wide range of different reinforcement learning algorithms. They all use the basic approach:

1. The agent analyzes the current state of the environment.
2. Takes the optimal action (within the framework of the learned Policy - behavior strategy).
3. Moves into a new state of the environment.
4. Receives a reward from the environment for a complete transition to a new state.

The sequence is based on the principles of the Markov process. It is assumed that the starting point is the current state of the environment. There is only one optimal way out of this state and it does not depend on the previous path.

more...

The financial markets industry is a complex and multifaceted environment. Every event and action have their roots in economic fundamental processes. The reason for certain events can be found in the news, geopolitical events, various technical aspects and many other factors. Quite often, we observe such dependencies after they happen. While analyzing the market situation, we observe only a small part of these factors. This in general makes financial markets a rather difficult environment to analyze. But still, we highlight some of the most significant tools that can detect the main trends. Other factors are attributed to environmental stochasticity.

In such a complex environment, reinforcement learning represents a powerful tool for developing strategies in financial markets. However, existing methods, such as Decision Transformer, may not be adaptive enough in highly stochastic environments. This is what we observed in the practical part of the previous article.

more...

Time series analysis plays an important role not just in supporting fundamental analysis but in very liquid markets like forex, it can be the main driver for decisions on how one is positioned in the markets. Traditional technical indicators have tended to lag the market a lot which has brought them out of favor for most traders, leading to the rise of alternatives perhaps the most predominant of which, at the moment is neural networks. But what about polynomial interpolation?

more...

LDA is a supervised generalization machine learning algorithm that aims to find a linear combination of features that best separates the classes in a dataset.


Just like the Principal Component Analysis(PCA), it is a dimension reduction algorithm, These algorithms are a common choice for dimensionality reduction, in this article we are going to compare them and observe in what situation each algorithm works best. We already discussed the PCA in the prior articles of this series, Let us commence by observing what the PCA algorithm is all about as we will discuss it mostly, finally we will compare their performances on a simple dataset and in the strategy tester, make sure you stick to the end for awesome data science stuff.

more...

The last two articles were devoted to the Decision Transformer method, which models action sequences in the context of an autoregressive model of desired rewards. As you might remember, according to the results of practical tests of two articles, the beginning of the testing period saw a fairly good increase in the profitability of the trained model results. Further on, the performance of the model decreases and a number of unprofitable transactions are observed, which leads to losses. The amount of losses received may exceed previously received profits.

more...