Articles on data analysis and statistics in MQL5

In this article, we will continue diving into the implementation of the ACMO (Atmospheric Cloud Model Optimization) algorithm. In particular, we will discuss two key aspects: the movement of clouds into low-pressure regions and the rain simulation, including the initialization of droplets and their distribution among clouds. We will also look at other methods that play an important role in managing the state of clouds and ensuring their interaction with the environment.

In this article, we create a canvas-based graphing tool in MQL5 for statistical correlation and linear regression analysis between two symbols, with draggable and resizable features. We incorporate ALGLIB for regression calculations, dynamic tick labels, data points, and a stats panel displaying slope, intercept, correlation, and R-squared. This interactive visualization aids in pair trading insights, supporting customizable themes, borders, and real-time updates on new bars

Learn how to read a CSV file in MQL5 and organize its trading data into dynamic arrays. This article shows step by step how to count file elements, store all data in a single array, and separate each column into dedicated arrays, laying the foundation for advanced analysis and trading performance visualization.

In this article, we will look at how to solve certain problems and issues that arise when using Python code within other programs. More specifically, we will demonstrate a common issue encountered when using Excel in conjunction with MetaTrader 5, although we will be using Python to facilitate this interaction. However, this implementation has a minor drawback. It does not occur in all cases, but only in certain specific situations. When it does happen, it is necessary to understand the cause. In today’s article, we will begin explaining how to resolve this issue.

The Camel Algorithm, developed in 2016, simulates the behavior of camels in the desert to solve optimization problems, taking into account temperature, supply, and endurance. This article also presents a modified version of the algorithm (CAm) with key improvements: the use of a Gaussian distribution in generating solutions and the optimization of the oasis effect parameters.

If everything seemed right to you up to this point, it means you're not really thinking about the long term, when you start developing applications. Over time you will no longer need to program new applications, you will just have to make them work together. So let's see how to finish assembling the mouse indicator.

The article extends our copula toolkit with mixed copulas implemented natively in MQL5. We construct Clayton–Frank–Gumbel and Clayton–Student–t–Gumbel mixtures, estimate them via EM, and enable sparsity control through SCAD with cross‑validation. Provided scripts tune hyperparameters, compare mixtures using information criteria, and save trained models. Practitioners can apply these components to capture asymmetric tail dependence and embed the selected model in indicators or Expert Advisors.

In this article, we look to explore how a complimentary indicator pairing can be used to analyze the recent 5-year history of Vanguard Information Technology Index Fund ETF. By considering two options of algorithms, Kendall’s Tau and Distance-Correlation, we look to select not just an ideal indicator pair for trading the VGT, but also suitable signal-pattern pairings of these two indicators.

This article presents the Chow test for detecting structural breaks in pair relationships and the application of the Cumulative Sum of Squares - CUSUM - for structural breaks monitoring and early detection. The article uses the Nvidia/Intel partnership announcement and the US Gov foreign trade tariff announcement as examples of slope inversion and intercept shift, respectively. Python scripts for all the tests are provided.

Integrating Python's logging module with MQL5 empowers traders with a systematic logging approach, simplifying the process of monitoring, debugging, and documenting trading activities. This article explains the adaptation process, offering traders a powerful tool for maintaining clarity and organization in trading software development.

This article presents a unique experiment that aims to examine the behavior of population optimization algorithms in the context of their ability to efficiently escape local minima when population diversity is low and reach global maxima. Working in this direction will provide further insight into which specific algorithms can successfully continue their search using coordinates set by the user as a starting point, and what factors influence their success.

We wrap up our look at learning rate sensitivity to the performance of Expert Advisors by primarily examining the Adaptive Learning Rates. These learning rates aim to be customized for each parameter in a layer during the training process and so we assess potential benefits vs the expected performance toll.

In this article, we expand the MQL5 graphing tool to support seventeen statistical distributions with interactive cycling via a header switch icon. We add type-specific data loading, discrete and continuous histogram computation, and theoretical density functions for each model, with dynamic titles, axis labels, and parameter panels that adapt automatically. The result lets you overlay distribution models on the same sample and compare fit across families without reloading the tool.

Create a CSV trading journal in MQL5 by reading account history over a defined period and writing structured records to file. The article explains deal counting, ticket retrieval, symbol and order type decoding, and capturing entry (lot, time, price, SL/TP) and exit (time, price, profit, result) data with dynamic arrays. The result is an organized, persistent log suitable for analysis and reporting.

We build a production MQL5 bet‑sizing toolkit: utilities, snippets, and user‑level functions that mirror the Python originals. The methods cover probability‑to‑size mapping with overlap correction, dynamic forecast‑price sizing (calibrated sigmoid/power with limit price), occupancy‑based budgeting, and mixture‑model reserve sizing (EF3M). The result is a signed [−1, ..., 1] position plus diagnostics you can plug directly into order logic.

Deterministic Oscillatory Search (DOS) algorithm is an innovative global optimization method that combines the advantages of gradient and swarm algorithms without the use of random numbers. The fitness oscillation and slope mechanism allows DOS to explore complex search spaces in a deterministic manner.