Articles on machine learning in trading

The article presents the Central Force Optimization (CFO) algorithm inspired by the laws of gravity. It explores how principles of physical attraction can solve optimization problems where "heavier" solutions attract less successful counterparts.

We are making a remote professional risk manager for Forex in Python, deploying it on the server step by step. In the course of the article, we will understand how to programmatically manage Forex risks, and how not to waste a Forex deposit any more.

We continue to explore the innovative Chimera framework – a two-dimensional state-space model that uses neural network technologies to analyze multidimensional time series. This method provides high forecasting accuracy with low computational cost.

How does the market observe Fibonacci-based relationships? This sequence, where each subsequent number is equal to the sum of the two previous ones (1, 1, 2, 3, 5, 8, 13, 21...), not only describes the growth of the rabbit population. We will consider the Pythagorean hypothesis that everything in the world is subject to certain relationships of numbers...

A new bioinspired optimization metaheuristic, NOA (Neuroboids Optimization Algorithm), combines the principles of collective intelligence and neural networks. Unlike conventional methods, the algorithm uses a population of self-learning "neuroboids", each with its own neural network that adapts its search strategy in real time. The article reveals the architecture of the algorithm, the mechanisms of self-learning of agents, and the prospects for applying this hybrid approach to complex optimization problems.

In this article, we present an MQL5 library for modeling volatility, designed to function similarly to Python's arch package. The library currently supports the specification of common conditional mean (HAR, AR, Constant Mean, Zero Mean) and conditional volatility (Constant Variance, ARCH, GARCH) models.

Successful Restaurateur Algorithm (SRA) is an innovative optimization method inspired by restaurant business management principles. Unlike traditional approaches, SRA does not discard weak solutions, but improves them by combining with elements of successful ones. The algorithm shows competitive results and offers a fresh perspective on balancing exploration and exploitation in optimization problems.

We continue exploring a multi-task learning framework based on ResNeXt, which is characterized by modularity, high computational efficiency, and the ability to identify stable patterns in data. Using a single encoder and specialized "heads" reduces the risk of model overfitting and improves the quality of forecasts.

In this article, we will attempt to predict the market with a decent model for time series forecasting named DeepAR. A model that is a combination of deep neural networks and autoregressive properties found in models like ARIMA and Vector Autoregressive (VAR).

The BOA method is inspired by the classic game of billiards and simulates the search for optimal solutions as a game with balls trying to fall into pockets representing the best results. In this article, we will consider the basics of BOA, its mathematical model, and its efficiency in solving various optimization problems.

This article demonstrates how the stochastic oscillator, a classical technical indicator, can be repurposed beyond its conventional use as a mean-reversion tool. By viewing the indicator through a different analytical lens, we show how familiar strategies can yield new value and support alternative trading rules, including trend-following interpretations. Ultimately, the article highlights how every technical indicator in the MetaTrader 5 terminal holds untapped potential, and how thoughtful trial and error can uncover meaningful interpretations hidden from view.

MQL5 lacks built-in asymmetric cryptography, making secure data exchange over insecure channels like HTTP difficult. This article presents a pure MQL5 implementation of RSA using PKCS#1 v1.5 padding, enabling safe transmission of AES session keys and small data blocks without external libraries. This approach provides HTTPS-like security over standard HTTP and even more, it fills an important gap in secure communication for MQL5 applications.

The article presents a new metaheuristic algorithm, Chaos Game Optimization (CGO), which demonstrates a unique ability to maintain high efficiency when dealing with high-dimensional problems. Unlike most optimization algorithms, CGO not only does not lose, but sometimes even increases performance when scaling a problem, which is its key feature.

A multi-task learning framework based on ResNeXt optimizes the analysis of financial data, taking into account its high dimensionality, nonlinearity, and time dependencies. The use of group convolution and specialized heads allows the model to effectively extract key features from the input data.

Tired of watching progress bars instead of testing trading strategies? Traditional caching fails financial ML, leaving you with lost computations and frustrating restarts. We've engineered a sophisticated caching architecture that understands the unique challenges of financial data—temporal dependencies, complex data structures, and the constant threat of look-ahead bias. Our three-layer system delivers dramatic speed improvements while automatically invalidating stale results and preventing costly data leaks. Stop waiting for computations and start iterating at the pace the markets demand.

We continue to build the Hidformer hierarchical dual-tower transformer model designed for analyzing and forecasting complex multivariate time series. In this article, we will bring the work we started earlier to its logical conclusion — we will test the model on real historical data.

This article shows how to configure a black-box model to automatically uncover strong trading strategies using a data-driven approach. By using Mutual Information to prioritize the most learnable signals, we can build smarter and more adaptive models that outperform conventional methods. Readers will also learn to avoid common pitfalls like overreliance on surface-level metrics, and instead develop strategies rooted in meaningful statistical insight.

Many people, especially non=programmers, find it very difficult to transfer information between MetaTrader 5 and other programs. One such program is Excel. Many use Excel as a way to manage and maintain their risk control. It is an excellent program and easy to learn, even for those who are not VBA programmers. Here we will look at how to establish a connection between MetaTrader 5 and Excel (a very simple method).

In this part, we focus on how to merge real-time market feedback—such as live trade outcomes, volatility changes, and liquidity shifts—with adaptive model learning to maintain a responsive and self-improving trading system.

I present to you my new population optimization algorithm - Blood Inheritance Optimization (BIO), inspired by the human blood group inheritance system. In this algorithm, each solution has its own "blood type" that determines the way it evolves. Just as in nature where a child's blood type is inherited according to specific rules, in BIO new solutions acquire their characteristics through a system of inheritance and mutations.

The MacroHFT framework for high-frequency cryptocurrency trading uses context-aware reinforcement learning and memory to adapt to dynamic market conditions. At the end of this article, we will test the implemented approaches on real historical data to assess their effectiveness.

We will use a quantum computer from IBM to discover all price movement options. Sounds like science fiction? Welcome to the world of quantum computing for trading!

This article helps new community members search for and discover their own candlestick patterns. Describing these patterns can be daunting, as it requires manually searching and creatively identifying improvements. Here, we introduce the engulfing candlestick pattern and show how it can be enhanced for more profitable trading applications.

I invite you to explore the MacroHFT framework, which applies context-aware reinforcement learning and memory to improve high-frequency cryptocurrency trading decisions using macroeconomic data and adaptive agents.

The article presents a new metaheuristic optimization Circle Search Algorithm (CSA) based on the geometric properties of a circle. The algorithm uses the principle of moving points along tangents to find the optimal solution, combining the phases of global exploration and local exploitation.

We continue to implement the approaches proposed by the authors of the FinCon framework. FinCon is a multi-agent system based on Large Language Models (LLMs). Today, we will implement the necessary modules and conduct comprehensive testing of the model on real historical data.

We invite you to explore the FinCon framework, which is a a Large Language Model (LLM)-based multi-agent system. The framework uses conceptual verbal reinforcement to improve decision making and risk management, enabling effective performance on a variety of financial tasks.

Discover how to fix a critical flaw in financial machine learning that causes overfit models and poor live performance—label concurrency. When using the triple-barrier method, your training labels overlap in time, violating the core IID assumption of most ML algorithms. This article provides a hands-on solution through sample weighting. You will learn how to quantify temporal overlap between trading signals, calculate sample weights that reflect each observation's unique information, and implement these weights in scikit-learn to build more robust classifiers. Learning these essential techniques will make your trading models more robust, reliable and profitable.

In this discussion, we contrast the classical approach to time series cross-validation with modern alternatives that challenge its core assumptions. We expose key blind spots in the traditional method—especially its failure to account for evolving market conditions. To address these gaps, we introduce Effective Memory Cross-Validation (EMCV), a domain-aware approach that questions the long-held belief that more historical data always improves performance.