Articles with examples of trading robots developed in MQL5

The LSEAttention framework offers improvements to the Transformer architecture. It was designed specifically for long-term multivariate time series forecasting. The approaches proposed by the authors of the method can be applied to solve problems of entropy collapse and learning instability, which are often encountered with vanilla Transformer.

In this article, we develop an informational dashboard in MQL5 for monitoring multi-symbol positions and account metrics like balance, equity, and free margin. We implement a sortable grid with real-time updates, CSV export, and a glowing header effect to enhance usability and visual appeal.

Hybrid graph sequence models (GSM++) combine the advantages of different architectures to provide high-fidelity data analysis and optimized computational costs. These models adapt effectively to dynamic market data, improving the presentation and processing of financial information.

In this series of articles, we look at the challenges faced by algorithmic traders when deploying machine-learning-powered trading strategies. Some challenges within our community remain unseen because they demand deeper technical understanding. Today’s discussion acts as a springboard toward examining the blind spots of cross-validation in machine learning. Although often treated as routine, this step can easily produce misleading or suboptimal results if handled carelessly. This article briefly revisits the essentials of time series cross-validation to prepare us for more in-depth insight into its hidden blind spots.

Discover how to effortlessly import and utilize the History Manager EX5 library in your MQL5 source code to process trade histories in your MetaTrader 5 account in this series' final article. With simple one-line function calls in MQL5, you can efficiently manage and analyze your trading data. Additionally, you will learn how to create different trade history analytics scripts and develop a price-based Expert Advisor as practical use-case examples. The example EA leverages price data and the History Manager EX5 library to make informed trading decisions, adjust trade volumes, and implement recovery strategies based on previously closed trades.

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.

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.

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 continue to develop the algorithms for FinAgent, a multimodal financial trading agent designed to analyze multimodal market dynamics data and historical trading patterns.

Forecasting plays an important role in time series analysis. In the new article, we will talk about the benefits of time series patching.

News trading often requires managing multiple positions and symbols within a very short time due to heightened volatility. In today’s discussion, we address the challenges of multi-symbol trading by integrating this feature into our News Headline EA. Join us as we explore how algorithmic trading with MQL5 makes multi-symbol trading more efficient and powerful.

This project explores the fusion of deep learning and technical analysis to test trading strategies in forex. A Python script is used for rapid experimentation, employing an ONNX model alongside traditional indicators like PSAR, SMA, and RSI to predict EUR/USD movements. A MetaTrader 5 script then brings this strategy into a live environment, using historical data and technical analysis to make informed trading decisions. The backtesting results indicate a cautious yet consistent approach, with a focus on risk management and steady growth rather than aggressive profit-seeking.

How to use Renko bars with AI? Let's look at Renko trading on Forex with forecast accuracy of up to 59.27%. We will explore the benefits of Renko bars for filtering market noise, learn why volume is more important than price patterns, and how to set the optimal Renko block size for EURUSD. This is a step-by-step guide on integrating CatBoost, Python, and MetaTrader 5 to create your own Renko Forex forecasting system. It is ideal for traders looking to go beyond traditional technical analysis.

Having started developing a multi-currency EA, we have already achieved some results and managed to carry out several code improvement iterations. However, our EA was unable to work with pending orders and resume operation after the terminal restart. Let's add these features.

We are creating an adaptive self-learning trading expert advisor based on DQN machine learning, with multidimensional causal inference. The EA will successfully trade simultaneously on 7 currency pairs. And agents of different pairs will exchange information with each other.

Learn how to detect and display the Gartley harmonic pattern in MetaTrader 5 using MQL5. This article explains each step of the process, from identifying swing points to applying Fibonacci ratios and plotting the full pattern on the chart for clear visual confirmation.

In this article news filtration for individual news events based on their IDs will be implemented. In addition, previous SQL queries will be improved to provide additional information or reduce the query's runtime. Furthermore, the code built in the previous articles will be made functional.

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.

When working with time series, we always use the source data in their historical sequence. But is this the best option? There is an opinion that changing the sequence of the input data will improve the efficiency of the trained models. In this article I invite you to get acquainted with one of the methods for optimizing the input sequence.

In this article, we develop a 5 Drives pattern system in MQL5 that identifies bullish and bearish 5 Drives harmonic patterns using pivot points and Fibonacci ratios, executing trades with customizable entry, stop loss, and take-profit levels based on user-selected options. We enhance trader insight with visual feedback through chart objects like triangles, trendlines, and labels to clearly display the A-B-C-D-E-F pattern structure.

We build a calendar‑driven news filter entirely in MQL5, avoiding web requests and external DLLs. Part 1 covers loading and caching events, mapping them to symbols by currency, filtering by impact level, defining pre/post windows, and blocking new trades during active news, with optional pre‑news position closure. The result is a configurable, prop‑firm‑friendly control that reduces false pauses and protects entries during volatility.

This article explores a trading strategy that integrates Linear Discriminant Analysis (LDA) with Bollinger Bands, leveraging categorical zone predictions for strategic market entry signals.

Can chaos theory be applied to financial markets? In this article, we will consider how conventional Chaos theory and chaotic systems are different from the concept proposed by Bill Williams.

We invite you to get acquainted with the Hierarchical Vector Transformer (HiVT) method, which was developed for fast and accurate forecasting of multimodal time series.

Imagine transforming the traditional EA or indicator input properties into a real-time, on-chart control interface. This discussion builds upon our foundational work in the Market Periods Synchronizer indicator, marking a significant evolution in how we visualize and manage higher-timeframe (HTF) market structures. Here, we turn that concept into a fully interactive utility—a dashboard that brings dynamic control and enhanced multi-period price action visualization directly onto the chart. Join us as we explore how this innovation reshapes the way traders interact with their tools.

In this session, we will build a sophisticated, multi-signal Expert Advisor using the MQL5 Standard Library. This approach allows us to seamlessly blend built-in signals with our own custom logic, demonstrating how to construct a powerful and flexible trading algorithm. For more, click to read further.

An empirical study of Larry Williams' short-term trading patterns, showing how classic setups can be automated in MQL5, tested on real market data, and evaluated for consistency, profitability, and practical trading value.

In this article, we will make an attempt to develop our own trading strategy. Any trading strategy must be based on some kind of statistical advantage. Moreover, this advantage should exist for a long time.

An empirical study of Larry Williams' short-term trading patterns, showing how classic setups can be automated in MQL5, tested on real market data, and evaluated for consistency, profitability, and practical trading value.

In the previous parts, the Expert Advisor (EA) under development was able to use only a fixed position size for trading. This is acceptable for testing, but is not advisable when trading on a real account. Let's make it possible to trade using variable position sizes.

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.