Here we will prepare the ground so that if we need to add new functions to the code, this will happen smoothly and easily. The current code cannot yet cover or handle some of the things that will be necessary to make meaningful progress. We need everything to be structured in order to enable the implementation of certain things with the minimal effort. If we do everything correctly, we can get a truly universal system that can very easily adapt to any situation that needs to be handled.
We will be creating a simple hedge EA as a base for our more advanced Grid-Hedge EA, which will be a mixture of classic grid and classic hedge strategies. By the end of this article, you will know how to create a simple hedge strategy, and you will also get to know what people say about whether this strategy is truly 100% profitable.
Obviously the current metrics are very far from the ideal time for creating a 1-minute bar. That's the first thing we are going to fix. Fixing the synchronization problem is not difficult. This may seem hard, but it's actually quite simple. We did not make the required correction in the previous article since its purpose was to explain how to transfer the tick data that was used to create the 1-minute bars on the chart into the Market Watch window.
A new article from Design Patterns articles and we will take a look at one of its types which is behavioral patterns to understand how we can build communication methods between created objects effectively. By completing these Behavior patterns we will be able to understand how we can create and build a reusable, extendable, tested software.
In this article, we will analyse the impact of dividend announcements on stock market returns and see how investors can earn more returns than those offered by the market when they expect a company to announce dividends. In doing so, we will also check the validity of the Efficient Market Hypothesis in the context of the Indian Stock Market.
We have already talked more than once about the importance of correctly selecting the reward function, which we use to stimulate the desired behavior of the Agent by adding rewards or penalties for individual actions. But the question remains open about the decryption of our signals by the Agent. In this article, we will talk about reward decomposition in terms of transmitting individual signals to the trained Agent.
Dive into the fascinating realm of algorithmic trading with our beginner-friendly guide to MQL5 programming. Discover the essentials of MQL5, the language powering MetaTrader 5, as we demystify the world of automated trading. From understanding the basics to taking your first steps in coding, this article is your key to unlocking the potential of algorithmic trading even without a programming background. Join us on a journey where simplicity meets sophistication in the exciting universe of MQL5.
As the model is trained based on the experience reproduction buffer, the current Actor policy moves further and further away from the stored examples, which reduces the efficiency of training the model as a whole. In this article, we will look at the algorithm of improving the efficiency of using samples in reinforcement learning algorithms.
We consider XLY, SPDR’s consumer discretionary spending ETF and see if with tools in MetaTrader’s IDE we can sift through an array of data sets in selecting what could work with a forecasting model with a forward outlook of not more than a year.
In this article we will complete the development of a simulator for our system. The main goal here will be to configure the algorithm discussed in the previous article. This algorithm aims to create a RANDOM WALK movement. Therefore, to understand today's material, it is necessary to understand the content of previous articles. If you have not followed the development of the simulator, I advise you to read this sequence from the very beginning. Otherwise, you may get confused about what will be explained here.
The article considers handling trade request structures, namely creating a request, its preliminary verification before sending it to the server, the server's response to a trade request and the structure of trade transactions. We will create simple and convenient functions for sending trading orders to the server and, based on everything discussed, create an EA informing of trade transactions.
In this article we will continue the simulator development stage. this time we will see how to effectively create a RANDOM WALK type movement. This type of movement is very intriguing because it forms the basis of everything that happens in the capital market. In addition, we will begin to understand some concepts that are fundamental to those conducting market analysis.
In this article, we will see how to develop a quality score that your Expert Advisor can display in the strategy tester. We will look at two well-known calculation methods – Van Tharp and Sunny Harris.
Here we will simplify a few elements related to the work in the next article. I'll also explain how you can visualize what the simulator generates in terms of randomness.
In this article, we will look at how to lock the indicator while simply using the MQL5 language, and we will do it in a very interesting and amazing way.
In this article we present the implementation of Combinatorially Symmetric Cross Validation in pure MQL5, to measure the degree to which a overfitting may occure after optimizing a strategy using the slow complete algorithm of the Strategy Tester.
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.
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.
XLV is SPDR healthcare ETF and in an age where it is common to be bombarded by a wide array of traditional news items plus social media feeds, it can be pressing to select a data set for use with a model. We try to tackle this problem for this ETF by sizing up some of its critical data sets in MQL5.
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.
Within the framework of the engineering approach developed by the author based on the probability theory, the conditions for opening a profitable position are found and the optimal (profit-maximizing) take profit and stop loss values are calculated.
Developing a simulator can be much more interesting than it seems. Today we'll take a few more steps in this direction because things are getting more interesting.
In order to use the data that forms the bars, we must abandon replay and start developing a simulator. We will use 1 minute bars because they offer the least amount of difficulty.
Several fellow traders sent emails or commented about how to use this Multi-Currency EA on brokers with symbol names that have prefixes and/or suffixes, and also how to implement trading time zones or trading time sessions on this Multi-Currency EA.
This article, the final in our series to tackle functors as a subject, revisits monoids as a category. Monoids which we have already introduced in these series are used here to aid in position sizing, together with multi-layer perceptrons.
In this article, we will explore the application of regression models from the Scikit-learn package, attempt to convert them into ONNX format, and use the resultant models within MQL5 programs. Additionally, we will compare the accuracy of the original models with their ONNX versions for both float and double precision. Furthermore, we will examine the ONNX representation of regression models, aiming to provide a better understanding of their internal structure and operational principles.
We continue our discussion of reinforcement learning algorithms for solving continuous action space problems. In this article, I will present the Soft Actor-Critic (SAC) algorithm. The main advantage of SAC is the ability to find optimal policies that not only maximize the expected reward, but also have maximum entropy (diversity) of actions.
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, we expand the range of tasks of our agent. The training process will include some aspects of money and risk management, which are an integral part of any trading strategy.
In this article, we will have a look at yet another reinforcement learning approach. It is called goal-conditioned reinforcement learning (GCRL). In this approach, an agent is trained to achieve different goals in specific scenarios.
Training useful skills without an explicit reward function is one of the main challenges in hierarchical reinforcement learning. Previously, we already got acquainted with two algorithms for solving this problem. But the question of the completeness of environmental research remains open. This article demonstrates a different approach to skill training, the use of which directly depends on the current state of the system.
In the previous article, we introduced the DIAYN method, which offers the algorithm for learning a variety of skills. The acquired skills can be used for various tasks. But such skills can be quite unpredictable, which can make them difficult to use. In this article, we will look at an algorithm for learning predictable skills.
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.
In the context of reinforcement learning, model procrastination can be caused by several reasons. The article considers some of the possible causes of model procrastination and methods for overcoming them.
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.
In this article we present an algorithm for permuting price bars and detail how permutation tests can be used to recognize instances where strategy performance has been fabricated to deceive potential buyers of Expert Advisors.
The article describes hierarchical training models that offer an effective approach to solving complex machine learning problems. Hierarchical models consist of several levels, each of which is responsible for different aspects of the task.
This article discusses the use of the Go-Explore algorithm over a long training period, since the random action selection strategy may not lead to a profitable pass as training time increases.
