An EA can detect a wrong ONNX model in OnInit, before its first prediction, instead of failing or returning other probabilities in OnTick. COnnxContract checks the model's signature, features, class labels, and SHA-256 hash against a contract file. COnnxParity compares the terminal's probabilities with onnxruntime on a golden file of reference rows, and a Python tool builds both files from your data. The tool also audits the EA's prediction log, because a golden pass covers only its own rows.
Financial markets can shift between regimes, so detecting when the statistical behavior of returns changes is valuable for analysis. An MQL5 change point detector segments historical returns into periods with different mean and volatility using optimal partitioning with PELT pruning and a BIC-style default penalty. A sensitivity scan tests robustness by showing which detected breaks persist as settings vary.
The BAT framework transforms a chaotic stream of market data into accurate forecasts and well-informed trading decisions. Backtesting shows steady capital growth with controlled risks. The model's architecture is simple, scalable, and ready for further optimization.
CStopOutGuard shows how far a symbol's price can move before the modeled margin level reaches stop-out, and how much closer a planned order would bring that threshold. Positions are valued with OrderCalcProfit() and OrderCalcMargin() at each tested price instead of a fixed tick value, so volume margin tiers are included. Unit tests, test EAs, and a Python consistency check cover the search. The model keeps the spread and swaps at their current values, and CanAdd() does not replace OrderCheck().
Building a multi-symbol Expert Advisor — one that also trades using different strategies — is a real challenge even for an algorithmic trader with many years of experience. But with the introduction of a full-fledged built-in AI Assistant in the MetaTrader 5 terminal, this is now within the reach of developers of any skill level. You do not even need to be an expert in math or risk management — the language model will explain everything to you in the chat and write all the code for you.
We explore an alternative to neural network signal filters: association rules and basic probability theory in MQL5, without Python or external libraries. This guide demonstrates how to discretize market conditions (RSI, EMA, ATR, momentum, session), compile statistics on support, confidence, lift, and edge using Laplace smoothing, and apply the result as a transparent entry filter before placing a trade.
Group lines in a custom indicator take parameter slots in iCustom() and IndicatorCreate(), so numbers passed without placeholders silently shift to the wrong inputs. CIndicatorInputs reads the live slot layout, fills the group slots, rejects values the terminal would change, and verifies the handle on a chart, so an EA gets the requested settings or an error. In the Strategy Tester it builds the handle from a saved layout without verification. Unit tests and a tester EA show both modes.
We continue implementing the BAT framework in MQL5, showing how bidirectional correlation and the SATMA module enable the analysis of market dynamics in the context of the current state. Key architectural solutions that enable the framework to be adapted for financial data analysis are presented.
The article focuses on the Nawa–Furuhashi Bacterial Evolutionary Algorithm (BEA) and its two key operators: segmental bacterial mutation for local optimization and horizontal gene transfer for the exchange of successful solution fragments. We implement BEA within the C_AO framework as a finite-state machine with a fixed evaluation budget, test it on Hilly, Forest, and Megacity, and explain the terminological distinction from the Numaoka model.
We implement a Hedge (multiplicative-weights) ensemble in MQL5 that combines five simple experts: MA cross, RSI reversion, Bollinger and Donchian breakouts, and a MACD histogram flip. The EA updates weights bar by bar from realized direction, with an optional regret-based learning rate. We wire it into trade logic, logging weights and enforcing risk controls, then compare Hedge to equal-weight and single-expert modes across multiple symbols and timeframes to show how to tune η and evaluate impact.
A pause after a losing streak assumes that losses come in clusters. CRunsTest turns the closed trades of an account or a tester run into a sequence of wins and losses and checks it with the Wald-Wolfowitz runs test, streak checks against shuffled orders, and a drift test. No dependence was detected on the tested EAs. A power simulation shows which effects each check could see, and on the shorter test runs a "not detected" result carries little weight.
The article presents the BAT framework, which enables accurate and adaptive modeling of temporal dynamics. Using bidirectional temporal correlation, BAT transforms sequential changes in market data into structured, informative representations. The model combines high computational efficiency with the ability to integrate deeply into trading systems, enabling the identification of both short-term and long-term price movement patterns.
The article examines the theoretical foundations and practical implementation of the Gaussian mixture model in the MQL5 language. It introduces the CGMM class and provides examples of its use, as well as a methodology for selecting the optimal number of model components using the AIC and BIC information criteria. Considerable attention is paid to how the EM algorithm works — a recognized standard for training probabilistic models with hidden (latent) variables. Support for MAP training and parameter initialization using K-Means++ has been implemented. A ready-to-use tool for clustering, multivariate density estimation, and data generation is now available in MetaTrader 5 as well.
We present an adaptation of the E-STMFlow framework — a modern solution for building autonomous trading systems. In this article, we complete the implementation of the approaches proposed by the framework's authors. The test results demonstrate steady capital growth, minimal drawdowns, and a predictable distribution of risk, confirming the practical effectiveness of the approach and opening up opportunities for further optimization of the strategy.
There is a common misconception among the general public that a neural network or artificial intelligence can, in a sense, comprehend the world in which we live. In some cases, this might even be true. However, if you want to achieve faster convergence of a neural network—or even a single neuron—you need to understand that the scale of the values used often affects the convergence rate. In addition, in this article I'll show you something strange that happens in neural networks. It will blow your mind and make you think about a lot of things related to this topic.
The article shows how to move the GridSurvivor Expert Advisor’s MLP filter from offline training in Python to fully embedded training in MQL5. The network is trained on the history of the current symbol and timeframe, is periodically retrained, and is used as the final signal filter. This approach eliminates external files and normalization mismatches, making the Expert Advisor self-contained and reproducible.
The Chimp Optimization Algorithm (ChOA) mimics primate group hunting with a division of roles, and its binary variant, BChimp, applies this mechanism to feature selection problems. We implement a continuous core in C_AO; along the way, we identify and correct an inherited coefficient defect — one that is masked by binarization but disrupts the search in the continuous domain. The abstract provides a ready-made implementation and practical insights into the quality and robustness of the search.
The article explains the internal mechanics of the STSSM block and demonstrates how modern SSM approaches can be adapted to the event-driven logic of spiking models while maintaining high speed and expressive representations. Step by step, we work our way through the architecture, transforming the rigorous theory behind the authors' approach into a practical financial instrument for analyzing financial time series.
The third article in this series introduces a trainable feature graph in the Cellular10K architecture: the weights of the feature → feature connections are strengthened online after correct predictions and weakened after errors. The article covers soft initialization, the message-passing step, a local Hebbian learning rule, weight clipping, normalization, and decay. The article demonstrates integration with a cellular automaton and a binary predictor, along with diagnostics metrics and practical activation thresholds for controlling overfitting.
This article builds an AWS SigV4 signer in pure MQL5 and wraps it in a small S3 client for PutObject, GetObject, and ListObjectsV2. It explains canonicalization, HMAC‑SHA256, and key derivation, and shows how to handle empty‑payload hashing and GMT timestamps to avoid SignatureDoesNotMatch. A script uploads a model and a CSV, downloads them back, and compares SHA‑256 checksums, providing a practical, DLL‑free path to centralized artifact storage.
In the second part, the cellular automaton is transferred from a grid to a graph. The features become vertices in a graph with local and long-range small-world connections, and the cells become agents that interact not only with their geometric neighbors but also with their semantic neighbors. The article discusses graph-based feature filtering, the construction of a neighbor graph, updated coherence-based voting, and the Graph Coherence and Graph Health metrics. This reduces the impact of isolated outliers and accelerates the propagation of market regimes while maintaining full compatibility with MQL5.
CTradePreflight checks a trade request before OrderSend(), so an EA finds an off-step volume, a wrong-side stop, missing margin or a closed session instead of getting a server rejection. The class repairs only safe cases, reads the OrderCheck() retcode, and counts every refusal by reason. Unit tests and a random-delay stress EA show that levels placed very close to the price can still fail, and an optional delay buffer stops those rejections at the cost of blocking some valid requests.
In this article, the perceptron is presented as the basis of a neural network, and its implementation in MQL5 is described in detail. We will explain what activation functions and their derivatives are, the difference between forward propagation and backpropagation, and how the least squares loss function and the gradient are used. You will learn how to train weights and biases, verify the results using test scripts, and adjust the learning rate to achieve stable convergence.
This article explains why and how the gradient is used in training a perceptron, based on the least squares method and the chain rule for computing partial derivatives. We will demonstrate how to implement gradient calculation in the C_Neuron class using the MQL5 language and test it using examples with 1 and 2 inputs. You'll learn how to adjust the weights and bias using gradient descent and prepare for forward propagation and backpropagation.
The SDformerFlow framework transforms complex events in financial time series into structured representations, enabling the model to identify local fluctuations and global trends simultaneously. The multi-level U-structure ensures consistency between the forward pass and the backward pass, synchronization of gradients, and computational stability. Ultimately, SDformerFlow proves to be a powerful and flexible tool for building modern trading systems.
The classic SuperTrend loses accuracy when the market regime changes because it uses fixed ATR parameters and a fixed multiplier. The article examines the architecture of ML SuperTrend Pro v2.00 written entirely in MQL5: a background test matrix with adaptive parameter updates, a regime grid as a context detector, an accuracy layer consisting of five filters, and Parabolic-style rendering with well-designed buffers. The article presents the principles of L1 regularization, the results of a comparison with the classic SuperTrend, and practical recommendations for deployment and integration via iCustom.
Introducing an MQL5 implementation of the Coyote Optimization Algorithm: packs with local alphas, a median-based tendency, and built-in crossover enable parallel exploration of regions of the search space while keeping premature convergence in check. The algorithm is integrated into C_AO and has been tested on the standard test bench and in a composite anti-cheat test. The article includes code, pseudocode, and an analysis of the operators that enable the use of COA to optimize the parameters of a trading system.
We estimate the probability, at broker time 08:00, that EURUSD's 08:00–16:00 range exceeds 1.25 times the mean of the previous 20 complete, valid target ranges. A two-feature SPSS logistic regression model employs the completed 00:00–08:00 H1 window. Coefficients from 2020–2023 sessions are fixed for the 2024 holdout comparison against constant and 60-outcome rolling frequency baselines, using Brier scores and log losses. MQL5 scripts allow readers to extract broker data and recreate the probabilities in MetaTrader 5.
This article presents a live MQL5 indicator for monitoring portfolio risk concentration through eigenvalue decomposition. The indicator calculates rolling covariance statistics for three instruments using persistent prefix sums, aligns price data by timestamp, handles missing observations, and plots the dominant eigenvalue as a percentage of total portfolio variance. The implementation provides a practical way to observe changes in portfolio risk concentration as the covariance structure evolves over time.
We invite you to explore the E-STMFlow framework, which effectively processes event streams by extracting informative embeddings, filtering out noise, and identifying key movements. Its architecture makes it possible to identify complex relationships between features and provides scalability, accuracy, and high computational efficiency for intelligent analysis and forecasting.
We describe and implement CVO: infection as candidate generation, coordinate-wise normal perturbation, and a dynamic population. The algorithm is integrated into C_AO and has been tested on a standard benchmark. The analysis identifies a scaling-related cause of stagnation and provides a practical solution — a shift to a relative step based on the range width. The code is ready for use.
An EA that passes CodeBase validation can still fail under execution delay and a freeze level. Using a tick scalper, the article reproduces four such failures in the Strategy Tester and shows each fix with its limits. An ablation under a fixed delay measures each change on its own. The burst-based stop distance removes most rejected modifications, while the weekend case behind the session buffer did not recur in those runs.
This article implements a PSI-based drift watchdog for MQL5 EAs. It compares a feature and the model's predicted probability against a quantile-based reference using a rolling window at a configurable cadence, applies hysteresis to avoid flapping, and raises a webhook on state transitions. You get tested math, an audit CSV and a plotting indicator, plus tuning guidance for thresholds, window size, and cooldown.
This is, without a doubt, the article I've enjoyed writing the most so far. In it, I showed that we don't really need anything special to achieve a certain goal. We can achieve this in various ways, provided we have the necessary knowledge and are willing to learn and dedicate ourselves to the task. I would like to express my heartfelt thanks to everyone who helped me choose these functions. The material presented here is intended solely for educational purposes. Under no circumstances should this be viewed as intended for anything other than teaching and studying the concepts presented.
The article presents an adaptation of the SDformerFlow framework that achieves high adaptability by integrating spiking attention with multi-window convolution and weighted summation of query elements. The architecture allows each attention head to learn its own parameters, which improves the model's accuracy and sensitivity to the structure of the data being analyzed.
The article describes TradeMux as a bridge between a Python pipeline and the MetaTrader 5 terminal for the seamless transfer of trading decisions without duplicating logic. The article examines a four-layer production architecture and a complete Python execution service: connection setup, reading account and position data, signal generation (including CatBoost), pre-trade risk control, kill_switch, and a supervisor. Practical benefits include cross-broker normalization (RoboForex, IC Markets, Alpari, OANDA) and scaling from a single account to multi-account broadcasting without changing the trading logic.
A candidate for our ranking table is Beluga Whale Optimization, a metaheuristic based on three behavioral patterns of the beluga whale: pair swimming, hunting using Lévy flight, and population renewal through whale fall. During implementation, we discovered that the algorithm does not so much optimize as read the geometry of the test bench; we examine how this works and develop an honest, promising modification called BWOm.
This article implements a multi‑timeframe Granger causality monitor for MetaTrader 5. A Jupyter notebook tests predefined timeframe pairs, selects lags, applies multiple‑testing control, and exports a calibration table. An MQL5 Expert Advisor reloads the setup, recomputes the F‑test live on rolling windows, and marks pairs VALID/INVALID to gate its vote‑based BUY/SELL generator. The workflow shows alignment rules, live/offline parity, and a filtered vs unfiltered comparison on EURUSD.
We will create an indicator written entirely in MQL5 to display the activation function and its derivative on the chart, using ReLU as an example. We will explain how derivatives affect the choice of activation function and what precautions are needed at nondifferentiable points. Readers will be able to interactively visualize the curves and gain a practical foundation for deciding when to use certain activation functions and when not to.
The article presents an adaptation of the SDformerFlow spiking architecture for the task of dense analysis of price micro-movements. The spatio-temporal structure provides a high level of detail, while the spiking logic ensures computational efficiency and the ability to work with sparse, pulse-like data. As a result, traders gain access to a tool that detects even the slightest shifts in liquidity and provides the foundation for more accurate and consistent real-time decisions.