Articles on the MQL5 programming and use of trading robots

In this article, we will look at the use of a trailing stop in trading. We will assess how useful and effective it is, and how it can be used. The efficiency of a trailing stop largely depends on price volatility and the selection of the stop loss level. A variety of approaches can be used to set a stop loss.

Data mining is crucial to a data scientist and a trader because very often, the data isn't as straightforward as we think it is. The human eye can not understand the minor underlying pattern and relationships in the dataset, maybe the K-means algorithm can help us with that. Let's find out...

Models are trained offline using data from a prepared training dataset. While providing certain advantages, its negative side is that information about the environment is greatly compressed to the size of the training dataset. Which, in turn, limits the possibilities of exploration. In this article, we will consider a method that enables the filling of a training dataset with the most diverse data possible.

We will make the order system more flexible. Here we will consider changes to the code that will make it more flexible, which will allow us to change position stop levels much faster.

We continue to study unsupervised learning algorithms. Some readers might have questions regarding the relevance of recent publications to the topic of neural networks. In this new article, we get back to studying neural networks.

In the previous article, we implemented the Soft Actor-Critic algorithm, but were unable to train a profitable model. Here we will optimize the previously created model to obtain the desired results.

One of the key problems within reinforcement learning is environmental exploration. Previously, we have already seen the research method based on Intrinsic Curiosity. Today I propose to look at another algorithm: Exploration via Disagreement.

Let's move on to a more complete order system directly on the chart. In this article, I will show a way to fix the order system, or rather, to make it more intuitive.

The problem of reinforcement learning lies in the need to define a reward function. It can be complex or difficult to formalize. To address this problem, activity-based and environment-based approaches are being explored to learn skills without an explicit reward function.

There are methods that can be used to solve many problems that can be repeated. Once understand how to use these methods it can be very helpful to create your software effectively and apply the concept of DRY ((Do not Repeat Yourself). In this context, the topic of Design Patterns will serve very well because they are patterns that provide solutions to well-described and repeated problems.

We continue to study reinforcement learning algorithms. All the algorithms we have considered so far required the creation of a reward policy to enable the agent to evaluate each of its actions at each transition from one system state to another. However, this approach is rather artificial. In practice, there is some time lag between an action and a reward. In this article, we will get acquainted with a model training algorithm which can work with various time delays from the action to the reward.

In this article, I will use experimentation and non-standard approaches to develop a profitable trading system and check whether neural networks can be of any help for traders. MetaTrader 5 as a self-sufficient tool for using neural networks in trading. Simple explanation.

With the rapid development of artificial intelligence today, language models (LLMs) are an important part of artificial intelligence, so we should think about how to integrate powerful LLMs into our algorithmic trading. For most people, it is difficult to fine-tune these powerful models according to their needs, deploy them locally, and then apply them to algorithmic trading. This series of articles will take a step-by-step approach to achieve this goal.

The multi-currency expert advisor in this article is an expert advisor or automated trading that uses ZigZag indicator which are filtered with the Awesome Oscillator or filter each other's signals.

In the previous article, we introduced the DDPG method, which allows training models in a continuous action space. However, like other Q-learning methods, DDPG is prone to overestimating Q-function values. This problem often results in training an agent with a suboptimal strategy. In this article, we will look at some approaches to overcome the mentioned issue.

In this article, I will show a completely different approach to algorithmic trading I ended up with after quite a long time. Of course, all this has to do with my brute force program, which has undergone a number of changes that allow it to solve several problems simultaneously. Nevertheless, the article has turned out to be more general and as simple as possible, which is why it is also suitable for those who know nothing about brute force.

We continue to discuss algorithms for training trajectory prediction models. In this article, we will get acquainted with a method called "AutoBots".

Unlock the secrets of MQL5 programming in our latest article! Delve into the essentials of structures, classes, and time functions, empowering your coding journey. Whether you're a beginner or an experienced developer, our guide simplifies complex concepts, providing valuable insights for mastering MQL5. Elevate your programming skills and stay ahead in the world of algorithmic trading!

In this article, we will make the final step towards the EA's performance. So, be prepared for a long read. To make our Expert Advisor reliable, we will first remove everything from the code that is not part of the trading system.

The article discusses the practical aspects of using optimization algorithms to find the best EA parameters on the fly, as well as virtualization of trading operations and EA logic. The article can be used as an instruction for implementing optimization algorithms into an EA.

Today we will take our order system to the next level. But before that, we need to solve a few problems. Now we have some questions that are related to how we want to work and what things we do during the trading day.

In this article, we explored the classic grid strategy, detailing its automation using an Expert Advisor in MQL5 and analyzing initial backtest results. We highlighted the strategy's need for high holding capacity and outlined plans for optimizing key parameters like distance, takeProfit, and lot sizes in future installments. The series aims to enhance trading strategy efficiency and adaptability to different market conditions.

We continue studying distributed Q-learning algorithms. In previous articles, we have considered distributed and quantile Q-learning algorithms. In the first algorithm, we trained the probabilities of given ranges of values. In the second algorithm, we trained ranges with a given probability. In both of them, we used a priori knowledge of one distribution and trained another one. In this article, we will consider an algorithm which allows the model to train for both distributions.

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. In this article, we will look at another optimization algorithm for this method.

We continue studying the environment in reinforcement learning models. And in this article we will look at another algorithm – Go-Explore, which allows you to effectively explore the environment at the model training stage.

Since the first articles devoted to reinforcement learning, we have in one way or another touched upon 2 problems: exploring the environment and determining the reward function. Recent articles have been devoted to the problem of exploration in offline learning. In this article, I would like to introduce you to an algorithm whose authors completely eliminated the reward function.

The last two articles considered the Soft Actor-Critic algorithm, which incorporates entropy regularization into the reward function. This approach balances environmental exploration and model exploitation, but it is only applicable to stochastic models. The current article proposes an alternative approach that is applicable to both stochastic and deterministic models.