Articles on the MQL5 programming and use of trading robots

We will guide you through the entire process of DL with python to make a GRU ONNX model, culminating in the creation of an Expert Advisor (EA) designed for trading, and subsequently comparing GRU model with LSTM model.

From this article on, start creating library functionality to work with price data. Today, create an object class which will store all price data which arrived with yet another tick.

This article describes one of the possible approaches to data transformation aimed at improving the generalizability of the model, and also discusses sampling and selection of CatBoost models.

In the last article, we got acquainted with the Autoencoder algorithm. Like any other algorithm, it has its advantages and disadvantages. In its original implementation, the autoenctoder is used to separate the objects from the training sample as much as possible. This time we will talk about how to deal with some of its disadvantages.

The article describes a method of fast optimization using the particle swarm algorithm. It also presents the method implementation in MQL, which is ready for use both in single-threaded mode inside an Expert Advisor and in a parallel multi-threaded mode as an add-on that runs on local tester agents.

In this article, we create several classes to facilitate real-time communication between MQL5 and Telegram. We focus on retrieving commands from Telegram, decoding and interpreting them, and sending appropriate responses back. By the end, we ensure that these interactions are effectively tested and operational within the trading environment

We will walk around statistical arbitrage, we will search with python for correlation and cointegration symbols, we will make an indicator for Pearson's coefficient and we will make an EA for trading statistical arbitrage with predictions done with python and ONNX models.

We got acquainted with the Q-learning method in one of the earlier articles within this series. This method averages rewards for each action. Two works were presented in 2017, which show greater success when studying the reward distribution function. Let's consider the possibility of using such technology to solve our problems.

The best practices, defining how to safely us an indicator, are not always easy to follow. Quiet market conditions may surprisingly produce readings on the indicator that do not qualify as a trading signal, leading to missed opportunities for algorithmic traders. This article will suggest a potential solution to this problem, as we discuss how to build trading applications capable of adapting their trading rules to the available market data.

Trading is, first of all, a calculus of probabilities. The proverb about idleness being an engine for progress reveals us the reason why all those indicators and trading systems have been developed. It comes that the major of newcomers in trading study "ready-made" trading theories. But, as luck would have it, there are some more undiscovered market secrets, and tools used in analyzing of price movements exist, basically, as those unrealized technical indicators or math and stat packages. Thanks awfully to Bill Williams for his contribution to the market movements theory. Though, perhaps, it's too early to rest on oars.

In this article, we develop a ChatGPT-integrated program in MQL5 with a user interface, leveraging the JSON parsing framework from Part 1 to send prompts to OpenAI’s API and display responses on a MetaTrader 5 chart. We implement a dashboard with an input field, submit button, and response display, handling API communication and text wrapping for user interaction.