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
Causal inference in time series classification problems
In this article, we will look at the theory of causal inference using machine learning, as well as the custom approach implementation in Python. Causal inference and causal thinking have their roots in philosophy and psychology and play an important role in our understanding of reality.
Neural Networks Made Easy (Part 90): Frequency Interpolation of Time Series (FITS)
By studying the FEDformer method, we opened the door to the frequency domain of time series representation. In this new article, we will continue the topic we started. We will consider a method with which we can not only conduct an analysis, but also predict subsequent states in a particular area.
Neural Networks in Trading: Generalized 3D Referring Expression Segmentation
While analyzing the market situation, we divide it into separate segments, identifying key trends. However, traditional analysis methods often focus on one aspect and thus limit the proper perception. In this article, we will learn about a method that enables the selection of multiple objects to ensure a more comprehensive and multi-layered understanding of the situation.
An introduction to Receiver Operating Characteristic curves
ROC curves are graphical representations used to evaluate the performance of classifiers. Despite ROC graphs being relatively straightforward, there exist common misconceptions and pitfalls when using them in practice. This article aims to provide an introduction to ROC graphs as a tool for practitioners seeking to understand classifier performance evaluation.
Neural networks are easy (Part 59): Dichotomy of Control (DoC)
In the previous article, we got acquainted with the Decision Transformer. But the complex stochastic environment of the foreign exchange market did not allow us to fully implement the potential of the presented method. In this article, I will introduce an algorithm that is aimed at improving the performance of algorithms in stochastic environments.
Neural Network in Practice: Least Squares
In this article, we'll look at a few ideas, including how mathematical formulas are more complex in appearance than when implemented in code. In addition, we will consider how to set up a chart quadrant, as well as one interesting problem that may arise in your MQL5 code. Although, to be honest, I still don't quite understand how to explain it. Anyway, I'll show you how to fix it in code.
Category Theory in MQL5 (Part 17): Functors and Monoids
This article, the final in our series to tackle functors as a subject, revisits monoids as a category. Monoids which we have already introduced in these series are used here to aid in position sizing, together with multi-layer perceptrons.
Feature Engineering for ML (Part 1): Fractional Differentiation — Stationarity Without Memory Loss
Integer differentiation forces a binary choice between stationarity and memory: returns (d=1) are stationary but discard all price-level information; raw prices (d=0) preserve memory but violate ML stationarity assumptions. We implement the fixed-width fractional differentiation (FFD) method from AFML Chapter 5, covering get_weights_ffd (iterative recurrence with threshold cutoff), frac_diff_ffd (bounded dot product per bar), and fracdiff_optimal (binary search for minimum stationary d*).
Most notable Artificial Cooperative Search algorithm modifications (ACSm)
Here we will consider the evolution of the ACS algorithm: three modifications aimed at improving the convergence characteristics and the algorithm efficiency. Transformation of one of the leading optimization algorithms. From matrix modifications to revolutionary approaches regarding population formation.
Self Optimizing Expert Advisor With MQL5 And Python (Part VI): Taking Advantage of Deep Double Descent
Traditional machine learning teaches practitioners to be vigilant not to overfit their models. However, this ideology is being challenged by new insights published by diligent researches from Harvard, who have discovered that what appears to be overfitting may in some circumstances be the results of terminating your training procedures prematurely. We will demonstrate how we can use the ideas published in the research paper, to improve our use of AI in forecasting market returns.
Neural networks made easy (Part 82): Ordinary Differential Equation models (NeuralODE)
In this article, we will discuss another type of models that are aimed at studying the dynamics of the environmental state.
Overcoming The Limitation of Machine Learning (Part 7): Automatic Strategy Selection
This article demonstrates how to automatically identify potentially profitable trading strategies using MetaTrader 5. White-box solutions, powered by unsupervised matrix factorization, are faster to configure, more interpretable, and provide clear guidance on which strategies to retain. Black-box solutions, while more time-consuming, are better suited for complex market conditions that white-box approaches may not capture. Join us as we discuss how our trading strategies can help us carefully identify profitable strategies under any circumstance.
Data Science and ML (Part 44): Forex OHLC Time series Forecasting using Vector Autoregression (VAR)
Explore how Vector Autoregression (VAR) models can forecast Forex OHLC (Open, High, Low, and Close) time series data. This article covers VAR implementation, model training, and real-time forecasting in MetaTrader 5, helping traders analyze interdependent currency movements and improve their trading strategies.
Chemical reaction optimization (CRO) algorithm (Part I): Process chemistry in optimization
In the first part of this article, we will dive into the world of chemical reactions and discover a new approach to optimization! Chemical reaction optimization (CRO) uses principles derived from the laws of thermodynamics to achieve efficient results. We will reveal the secrets of decomposition, synthesis and other chemical processes that became the basis of this innovative method.
Overcoming The Limitation of Machine Learning (Part 2): Lack of Reproducibility
The article explores why trading results can differ significantly between brokers, even when using the same strategy and financial symbol, due to decentralized pricing and data discrepancies. The piece helps MQL5 developers understand why their products may receive mixed reviews on the MQL5 Marketplace, and urges developers to tailor their approaches to specific brokers to ensure transparent and reproducible outcomes. This could grow to become an important domain-bound best practice that will serve our community well if the practice were to be widely adopted.
Neural Networks in Trading: Hierarchical Dual-Tower Transformer (Final Part)
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.
Population optimization algorithms: Whale Optimization Algorithm (WOA)
Whale Optimization Algorithm (WOA) is a metaheuristic algorithm inspired by the behavior and hunting strategies of humpback whales. The main idea of WOA is to mimic the so-called "bubble-net" feeding method, in which whales create bubbles around prey and then attack it in a spiral motion.
African Buffalo Optimization (ABO)
The article presents the African Buffalo Optimization (ABO) algorithm, a metaheuristic approach developed in 2015 based on the unique behavior of these animals. The article describes in detail the stages of the algorithm implementation and its efficiency in finding solutions to complex problems, which makes it a valuable tool in the field of optimization.
MQL5 Wizard Techniques You Should Know (Part 15): Support Vector Machines with Newton's Polynomial
Support Vector Machines classify data based on predefined classes by exploring the effects of increasing its dimensionality. It is a supervised learning method that is fairly complex given its potential to deal with multi-dimensioned data. For this article we consider how it’s very basic implementation of 2-dimensioned data can be done more efficiently with Newton’s Polynomial when classifying price-action.
Battle Royale Optimizer (BRO)
The article explores the Battle Royale Optimizer algorithm — a metaheuristic in which solutions compete with their nearest neighbors, accumulate “damage,” are replaced when a threshold is exceeded, and periodically shrink the search space around the current best solution. It presents both pseudocode and an MQL5 implementation of the CAOBRO class, including neighbor search, movement toward the best solution, and an adaptive delta interval. Test results on the Hilly, Forest, and Megacity functions highlight the strengths and limitations of the approach. The reader is provided with a ready-to-use foundation for experimentation and tuning key parameters such as popSize and maxDamage.
Generative Adversarial Networks (GANs) for Synthetic Data in Financial Modeling (Part 2): Creating Synthetic Symbol for Testing
In this article we are creating a synthetic symbol using a Generative Adversarial Network (GAN) involves generating realistic Financial data that mimics the behavior of actual market instruments, such as EURUSD. The GAN model learns patterns and volatility from historical market data and creates synthetic price data with similar characteristics.
Ordinal Encoding for Nominal Variables
In this article, we discuss and demonstrate how to convert nominal predictors into numerical formats that are suitable for machine learning algorithms, using both Python and MQL5.
Category Theory in MQL5 (Part 12): Orders
This article which is part of a series that follows Category Theory implementation of Graphs in MQL5, delves in Orders. We examine how concepts of Order-Theory can support monoid sets in informing trade decisions by considering two major ordering types.
Hybridization of population algorithms. Sequential and parallel structures
Here we will dive into the world of hybridization of optimization algorithms by looking at three key types: strategy mixing, sequential and parallel hybridization. We will conduct a series of experiments combining and testing relevant optimization algorithms.
Brain Storm Optimization algorithm (Part I): Clustering
In this article, we will look at an innovative optimization method called BSO (Brain Storm Optimization) inspired by a natural phenomenon called "brainstorming". We will also discuss a new approach to solving multimodal optimization problems the BSO method applies. It allows finding multiple optimal solutions without the need to pre-determine the number of subpopulations. We will also consider the K-Means and K-Means++ clustering methods.
Building Volatility models in MQL5 (Part I): The Initial Implementation
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.
Mutual information as criteria for Stepwise Feature Selection
In this article, we present an MQL5 implementation of Stepwise Feature Selection based on the mutual information between an optimal predictor set and a target variable.
MQL5 Wizard Techniques you should know (Part 68): Using Patterns of TRIX and the Williams Percent Range with a Cosine Kernel Network
We follow up our last article, where we introduced the indicator pair of TRIX and Williams Percent Range, by considering how this indicator pairing could be extended with Machine Learning. TRIX and William’s Percent are a trend and support/ resistance complimentary pairing. Our machine learning approach uses a convolution neural network that engages the cosine kernel in its architecture when fine-tuning the forecasts of this indicator pairing. As always, this is done in a custom signal class file that works with the MQL5 wizard to assemble an Expert Advisor.
Applying Localized Feature Selection in Python and MQL5
This article explores a feature selection algorithm introduced in the paper 'Local Feature Selection for Data Classification' by Narges Armanfard et al. The algorithm is implemented in Python to build binary classifier models that can be integrated with MetaTrader 5 applications for inference.
Camel Algorithm (CA)
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.
Ensemble methods to enhance classification tasks in MQL5
In this article, we present the implementation of several ensemble classifiers in MQL5 and discuss their efficacy in varying situations.
Artificial Tribe Algorithm (ATA)
The article provides a detailed discussion of the key components and innovations of the ATA optimization algorithm, which is an evolutionary method with a unique dual behavior system that adapts depending on the situation. ATA combines individual and social learning while using crossover for explorations and migration to find solutions when stuck in local optima.
Chaos optimization algorithm (COA): Continued
We continue studying the chaotic optimization algorithm. The second part of the article deals with the practical aspects of the algorithm implementation, its testing and conclusions.
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.
Central Force Optimization (CFO) algorithm
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.
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.
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.
Neuroboids Optimization Algorithm 2 (NOA2)
The new proprietary optimization algorithm NOA2 (Neuroboids Optimization Algorithm 2) combines the principles of swarm intelligence with neural control. NOA2 combines the mechanics of a neuroboid swarm with an adaptive neural system that allows agents to self-correct their behavior while searching for the optimum. The algorithm is under active development and demonstrates potential for solving complex optimization problems.
Neural Networks in Trading: Multi-Task Learning Based on the ResNeXt Model (Final Part)
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.
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.