Articles on machine learning in trading
Creating AI-based trading robots: native integration with Python, matrices and vectors, math and statistics libraries and much more.
Find out how to use machine learning in trading. Neurons, perceptrons, convolutional and recurrent networks, predictive models — start with the basics and work your way up to developing your own AI. You will learn how to train and apply neural networks for algorithmic trading in financial markets.
Add a new article
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Analyzing weather impact on currencies of agricultural countries using Python
What is the relationship between weather and Forex? Classical economic theory has long ignored the influence of such factors as weather on market behavior. But everything has changed. Let's try to find connections between the weather conditions and the position of agricultural currencies on the market.
Feature Engineering for ML (Part 13): Trend-Scanning Features in Python
Trend-scanning supports both forward and backward windows, and the labeling default is unsafe for features: it looks ahead and boosts next-bar agreement well above chance on random walks. We provide a dedicated wrapper, get trend scanning features, that forces computational causal and returns only window, slope, t value, and rsquared. A second analysis quantifies errors introduced by the default log transform on signed series.
Analyzing binary code of prices on the exchange (Part I): A new look at technical analysis
This article presents an innovative approach to technical analysis based on converting price movements into binary code. The author demonstrates how various aspects of market behavior — from simple price movements to complex patterns — can be encoded in a sequence of zeros and ones.
Bivariate Copulae in MQL5 (Part 1): Implementing Gaussian and Student's t-Copulae for Dependency Modeling
This is the first part of an article series presenting the implementation of bivariate copulae in MQL5. This article presents code implementing Gaussian and Student's t-copulae. It also delves into the fundamentals of statistical copulae and related topics. The code is based on the Arbitragelab Python package by Hudson and Thames.
MQL5 Wizard Techniques you should know (Part 54): Reinforcement Learning with hybrid SAC and Tensors
Soft Actor Critic is a Reinforcement Learning algorithm that we looked at in a previous article, where we also introduced python and ONNX to these series as efficient approaches to training networks. We revisit the algorithm with the aim of exploiting tensors, computational graphs that are often exploited in Python.
Overcoming The Limitation of Machine Learning (Part 4): Overcoming Irreducible Error Using Multiple Forecast Horizons
Machine learning is often viewed through statistical or linear algebraic lenses, but this article emphasizes a geometric perspective of model predictions. It demonstrates that models do not truly approximate the target but rather map it onto a new coordinate system, creating an inherent misalignment that results in irreducible error. The article proposes that multi-step predictions, comparing the model’s forecasts across different horizons, offer a more effective approach than direct comparisons with the target. By applying this method to a trading model, the article demonstrates significant improvements in profitability and accuracy without changing the underlying model.
MetaTrader 5 Machine Learning Blueprint (Part 17): CPCV Backtesting — From Python Model to Tick-Level Evidence
We bridge Python-native artifacts to MQL5 for tick-accurate CPCV backtesting. The export script converts the ONNX model, calibrator, feature spec, and path masks to flat files, while the expert advisor rebuilds features, performs ONNX inference with calibration, and trades on real ticks. The Strategy Tester runs each combinatorial path, and Python aggregates per-path equities into a path Sharpe distribution to assess robustness after spread, slippage, and commission.
Role of random number generator quality in the efficiency of optimization algorithms
In this article, we will look at the Mersenne Twister random number generator and compare it with the standard one in MQL5. We will also find out the influence of the random number generator quality on the results of optimization algorithms.
Feature Engineering for ML (Part 2): Implementing Fixed-Width Fractional Differentiation in MQL5
This article delivers a production-grade MQL5 implementation of fixed-width fractional differentiation for live MetaTrader 5 feeds. We introduce a header-only CFFDEngine that precomputes weights without a fixed cap, performs O(width) per-bar updates, and avoids per-tick allocations. The FFD.mq5 indicator supports all ENUM_APPLIED_PRICE types and prev_calculated optimization. Validation scripts confirm numerical equivalence with the standard Python frac diff_ffd pipeline.
Neural Networks in Trading: Reducing Memory Consumption with Adam-mini Optimization
One of the directions for increasing the efficiency of the model training and convergence process is the improvement of optimization methods. Adam-mini is an adaptive optimization method designed to improve on the basic Adam algorithm.
Example of Causality Network Analysis (CNA) and Vector Auto-Regression Model for Market Event Prediction
This article presents a comprehensive guide to implementing a sophisticated trading system using Causality Network Analysis (CNA) and Vector Autoregression (VAR) in MQL5. It covers the theoretical background of these methods, provides detailed explanations of key functions in the trading algorithm, and includes example code for implementation.
MQL5 Wizard Techniques you should know (Part 47): Reinforcement Learning with Temporal Difference
Temporal Difference is another algorithm in reinforcement learning that updates Q-Values basing on the difference between predicted and actual rewards during agent training. It specifically dwells on updating Q-Values without minding their state-action pairing. We therefore look to see how to apply this, as we have with previous articles, in a wizard assembled Expert Advisor.
Population optimization algorithms: Evolution of Social Groups (ESG)
We will consider the principle of constructing multi-population algorithms. As an example of this type of algorithm, we will have a look at the new custom algorithm - Evolution of Social Groups (ESG). We will analyze the basic concepts, population interaction mechanisms and advantages of this algorithm, as well as examine its performance in optimization problems.
MQL5 Wizard Techniques you should know (14): Multi Objective Timeseries Forecasting with STF
Spatial Temporal Fusion which is using both ‘space’ and time metrics in modelling data is primarily useful in remote-sensing, and a host of other visual based activities in gaining a better understanding of our surroundings. Thanks to a published paper, we take a novel approach in using it by examining its potential to traders.
Market Microstructure in MQL5 (Part 1): Robust Foundation
This article builds the foundation layer of a twelve-part MQL5 market microstructure toolkit. It implements guarded math helpers (SafeDivide, SafeLog, SafeSqrt, SafeExp, SafeTanh), robust data validation (ValidateSymbolV2, SafeCopyClose), trimmed statistical estimators (robust mean var), a linear regression slope, shared structs, and an FFT. You compile a single include file that hardens indicators and expert advisors against silent numerical failures and standardizes data flow for later parts.
Feature Engineering for ML (Part 4): Implementing Time Features in MQL5
Applying Python session boundaries to MQL5 broker timestamps misclassifies session membership by two to three hours on any non-UTC broker, corrupting session flags across the full backtest history. We implement CTimeFeatures.mqh, containing CRingBuffer and CTimeFeatures, with three EA-facing methods: Initialize (UTC offset capture and frequency gate configuration), Update (log return push to session-conditional ring buffers), and Calculate (cyclical encoding, session flags, and session volatility). The output is a flat double array drop-compatible with Python's get_time_features for sub-hourly, hourly, and daily timeframes.
Detecting and Classifying Fractal Patterns Using Machine Learning
In this article, we will touch upon the intriguing topic of fractal analysis and market forecasting using machine learning. These are just the first steps towards exploring the diverse fractal structures that form on financial price charts. We will use the correlation to find patterns and the CatBoost algorithm to classify these patterns.
MQL5 Wizard Techniques you should know (Part 20): Symbolic Regression
Symbolic Regression is a form of regression that starts with minimal to no assumptions on what the underlying model that maps the sets of data under study would look like. Even though it can be implemented by Bayesian Methods or Neural Networks, we look at how an implementation with Genetic Algorithms can help customize an expert signal class usable in the MQL5 wizard.
Neural Networks in Trading: Two-Dimensional Connection Space Models (Chimera)
In this article, we will explore the innovative Chimera framework: a two-dimensional state-space model that uses neural networks to analyze multivariate time series. This method offers high accuracy with low computational cost, outperforming traditional approaches and Transformer architectures.
Bacterial Chemotaxis Optimization (BCO)
The article presents the original version of the Bacterial Chemotaxis Optimization (BCO) algorithm and its modified version. We will take a closer look at all the differences, with a special focus on the new version of BCOm, which simplifies the bacterial movement mechanism, reduces the dependence on positional history, and uses simpler math than the computationally heavy original version. We will also conduct the tests and summarize the results.
Population optimization algorithms: Charged System Search (CSS) algorithm
In this article, we will consider another optimization algorithm inspired by inanimate nature - Charged System Search (CSS) algorithm. The purpose of this article is to present a new optimization algorithm based on the principles of physics and mechanics.
Neuroboids Optimization Algorithm (NOA)
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.
MQL5 Wizard Techniques you should know (Part 81): Using Patterns of Ichimoku and the ADX-Wilder with Beta VAE Inference Learning
This piece follows up ‘Part-80’, where we examined the pairing of Ichimoku and the ADX under a Reinforcement Learning framework. We now shift focus to Inference Learning. Ichimoku and ADX are complimentary as already covered, however we are going to revisit the conclusions of the last article related to pipeline use. For our inference learning, we are using the Beta algorithm of a Variational Auto Encoder. We also stick with the implementation of a custom signal class designed for integration with the MQL5 Wizard.
Low-Frequency Quantitative Strategies in MetaTrader 5 (Part 5): Pre-Backtest Evaluation of Machine-Learning-Generated Signals Through Formulaic Alphas
The article shows how to evaluate machine-learning alphas before a full backtest by expressing them as formulaic alphas. We compute Information Coefficient (IC), Rank IC, Information Ratio (ICIR), and t-statistics to quantify forecasting strength and stability. A MetaTrader 5 backtest illustrates differences versus execution-dependent tests, and a Python parser facilitates reproducible calculations and bulk screening.
Neural network trading EA based on PatchTST
The article presents the revolutionary architecture of PatchTST, a tailored transformer for financial time series analysis that breaks market data into 16-bar patches for efficient processing. We will discuss the full implementation of a trading robot in MQL5 covering everything from mathematical fundamentals and data structures to a ready-made EA with risk management and continuous learning systems.
Neuro-Structural Trading Engine — NSTE (Part II): Jardine's Gate Six-Gate Quantum Filter
This article introduces Jardine's Gate, a six-gate orthogonal signal filter for MetaTrader 5 that validates LSTM predictions across entropy, expert interference, confidence, regime-adjusted probability, trend direction, and consecutive-loss kill switch dimensions. Out of 43,200 raw signals per month, only 127 pass all six gates. Readers get the complete QuantumEdgeFilter MQL5 class, threshold calibration logic, and gate performance analytics.
Gating mechanisms in ensemble learning
In this article, we continue our exploration of ensemble models by discussing the concept of gates, specifically how they may be useful in combining model outputs to enhance either prediction accuracy or model generalization.
Gaussian Processes in Machine Learning (Part 2): Implementing and Testing a Classification Model in MQL5
In this section, we will look at the implementation of the key interfaces of the library of Gaussian processes in MQL5: IKernel, ILikelihood, and IInference. We will also demonstrate its operation on synthetic data and implement indicators for classification and regression, demonstrating its operation in online mode - with retraining of the model on each new bar.
MQL5 Wizard Techniques you should know (Part 60): Inference Learning (Wasserstein-VAE) with Moving Average and Stochastic Oscillator Patterns
We wrap our look into the complementary pairing of the MA & Stochastic oscillator by examining what role inference-learning can play in a post supervised-learning & reinforcement-learning situation. There are clearly a multitude of ways one can choose to go about inference learning in this case, our approach, however, is to use variational auto encoders. We explore this in python before exporting our trained model by ONNX for use in a wizard assembled Expert Advisor in MetaTrader.
Analyzing Price Time Gaps in MQL5 (Part II): Creating a Heat Map of Liquidity Distribution Over Time
A detailed guide on how to create a heat map indicator for MetaTrader 5 that visualizes the price distribution over time. The article reveals the mathematical basis of time density analysis, where each price level is colored from red (minimum stay time) to blue (maximum stay time).
The Group Method of Data Handling: Implementing the Multilayered Iterative Algorithm in MQL5
In this article we describe the implementation of the Multilayered Iterative Algorithm of the Group Method of Data Handling in MQL5.
Population optimization algorithms: Binary Genetic Algorithm (BGA). Part I
In this article, we will explore various methods used in binary genetic and other population algorithms. We will look at the main components of the algorithm, such as selection, crossover and mutation, and their impact on the optimization. In addition, we will study data presentation methods and their impact on optimization results.
Neural Networks in Trading: LSTM Optimization for Multivariate Time Series Forecasting (Final Part)
We continue to implement the DA-CG-LSTM framework, which offers innovative methods for time series analysis and forecasting. The use of CG-LSTM and dual attention allows for more accurate detection of both long-term and short-term dependencies in data, which is particularly useful for working with financial markets.
Neural networks made easy (Part 63): Unsupervised Pretraining for Decision Transformer (PDT)
We continue to discuss the family of Decision Transformer methods. From previous article, we have already noticed that training the transformer underlying the architecture of these methods is a rather complex task and requires a large labeled dataset for training. In this article we will look at an algorithm for using unlabeled trajectories for preliminary model training.
MQL5 Wizard Techniques you should know (Part 35): Support Vector Regression
Support Vector Regression is an idealistic way of finding a function or ‘hyper-plane’ that best describes the relationship between two sets of data. We attempt to exploit this in time series forecasting within custom classes of the MQL5 wizard.
Population optimization algorithms: Artificial Multi-Social Search Objects (MSO)
This is a continuation of the previous article considering the idea of social groups. The article explores the evolution of social groups using movement and memory algorithms. The results will help to understand the evolution of social systems and apply them in optimization and search for solutions.
Neural Networks in Trading: Actor—Director—Critic
We invite you to explore the Actor-Director-Critic framework, which combines hierarchical learning and a multi-component architecture for creating adaptive trading strategies. In this article, we take a detailed look at how using the Director to classify the Actor's actions helps to effectively optimize trading decisions and improve the robustness of models in financial market conditions.
Measuring What Matters (Part 3): The Reconstruction Engine — Validating Risk Footprints with Matrix Algebra
This article performs a numerical verification of MQL5 eigendecomposition for a covariance matrix using the spectral theorem A = V Λ Vᵀ. It reconstructs the matrix with Diag(), Transpose(), and MatMul(), computes the residual and its Frobenius norm, and shows that deviations remain at floating‑point precision, with results printed to the Experts journal.
Quantization in machine learning (Part 2): Data preprocessing, table selection, training CatBoost models
The article considers the practical application of quantization in the construction of tree models. The methods for selecting quantum tables and data preprocessing are considered. No complex mathematical equations are used.
Reinforcement Learning Meets MetaTrader 5: A Complete Pipeline for Training, Validating and Honestly Evaluating a Gold Trading Bot
This article presents a complete RL trading pipeline for XAUUSD: a supervised signal baseline with triple-barrier labels, PPO training, purged walk-forward validation with embargo, multi-seed checks, and contract-guarded deployment with normalization. It includes runnable code for data validation, features, environment, training, and broker‑based reconciliation. The live demo over 763 closed trades showed no statistically significant edge, and the methods highlight where information and costs, not architecture, set performance limits.