A practical introduction to time series regression models: regression on a constant and simple regression with deterministic, exogenous, and endogenous regressors. It describes the key steps in the diagnostic process, including residual analysis and hypothesis testing, that are necessary for making informed trading decisions. MQL5 scripts for MetaTrader 5 are included; they implement tests and plots using real data.
The article examines the architecture of a committee consisting of 15 AI agents: ten analysts and four risk officers vote in three parallel phases, and the Chair records the final result. Isolated pair contexts with separate reputations are used for eight currency pairs. The dynamic vote threshold depends on the daily PnL targets. The Expert Advisor uses only the signal, SL and TP, which allows for evaluation pf the quality of its decisions without any additional mechanics.
In the "Every tick based on real ticks" mode, the Strategy Tester's journal reports the minutes for which real ticks were absent or discarded and says that every tick generation was used for them. We show where the journal records this, measure how far back usable tick history goes on two demo servers, and check the ticks from inside an Expert Advisor, so a test range can be chosen by its real-tick share.
The article describes a test of the hypothesis of randomness for quotes based on a chi-square statistic derived from the frequencies of overlapping s-chains. The article shows how to define discrete states and compare observed and expected frequencies in order to detect Markov memory in price increments. This approach helps distinguish structural dependence from noise and formalize the testing of trading hypotheses.
The article discusses the projection of EURUSD daily candlesticks onto the secp256k1 elliptic curve and the extraction of 96 features (EC+TA) to predict the direction of the next candlestick in CatBoost. The price mapping to the curve and the training pipeline on 2,000 D1 bars are shown; the full model achieves a test AUC of 0.6508, with EC features contributing 60.6%. The materials are suitable for reproduction in Python/MetaTrader 5.
A compact MetaTrader 5 library builds a synthetic market from agents: a minority game, a zero‑intelligence double auction, a stylized‑facts analyzer, and a parameter scanner. The result is a testbed with known answers, providing the control real price backtests lack. Each result is cross‑checked by the model's phase transition, a matching Python implementation, and detection of a planted edge.
This article presents an MQL5 trade-management engine that converts exit rules into proposal generators. A policy engine ranks all proposals by a configurable reason priority with documented tie-breakers. A broker-aware validator checks the candidates in that order and falls back to the next one on rejection, and an execution adapter sends the first accepted request. Reconciliation confirms the result on the position before the internal state changes. The engine uses R-multiples for thresholds, persists intratrade state, and records a decision trace for each managed position on every tick.
We present a multi‑symbol EA structure that moves symbol‑specific state into independently initialized CStrategyNode objects and manages them with CPortfolioManager. Using CArrayObj with FreeMode, Add(), At(), Delete(), Total(), and Clear(), the manager ensures safe ownership and isolation, while timer‑driven routing and completed‑bar gates keep processing deterministic. This lets you scale portfolios, control risk per node, and avoid duplicated event handling.
The article addresses online false discovery rate control for streaming signals, where fixed thresholds inflate noise discoveries. We port LORD3, LORD++ and SAFFRON into MQL5, construct a causal bootstrap p‑value source without a resolution floor, and verify correctness before market data. Dependence stress tests, a budget/level indicator and a two-arm Strategy Tester show how to configure, monitor and throttle streams to keep the acted-on noise fraction within target.
This stage connects structural validation with reference matching for double bottoms. The Expert Advisor scans completed history, filters pairs by the current structural rules, and compares normalized close and low profiles against validated samples from the library. Results that meet the similarity requirement appear in the panel for chart review and are saved to REFERENCE_MATCHES.csv.
Tick volume mostly reflects the time of day and is only complete when a bar closes. We build a per‑quote, live activity gauge from the waits between quote changes, remove the intraday rhythm, and model the remainder with the autoregressive conditional duration (ACD) model. We then test whether this activity measure improves short‑horizon volatility forecasts once live price volatility is already in the model.
Permutation tests that assume exchangeable observations miscalibrate feature screens for overlapping financial labels, yielding a tenfold false‑positive inflation on EURUSD. This article implements a calibrated three‑gate pipeline: ADF+KPSS for stationarity, uniform‑grid entropy with modal occupancy, and mutual information corrected via a joint permutation null and Westfall–Young max‑T. The ScreeningReport summarizes gate outcomes and effective test count so you can filter features reliably before fitting models.
A practical route from the option chain to probability levels inside MetaTrader 5. For GLD, SLV, SPY, and USO, the risk-neutral density implied by each ETF's options is recovered with the Breeden-Litzenberger identity and checked for static arbitrage, after the method itself is validated against an analytic lognormal. A Python feed publishes the percentiles; an MQL5 service rescales them onto a related symbol, draws levels, and alerts when price leaves the risk-neutral 5 to 95 percent band. Every reading carries its own diagnostics, and full source is included.
An LLM agent with SQL memory in MetaTrader 5 — an architecture that eliminates the main limitation of traditional LLM Expert Advisors (EAs): the loss of context between queries. Each agent decision is recorded in SQLite along with the market conditions at the time of the signal — RSI, MA alignment, volatility — and after the position is closed, the outcome is written to the database. When the next request comes in, the agent reviews its own history — similar past RSI conditions and the last three agent decisions — and only then analyzes the current market. The result is a system that remembers its errors between restarts and adapts its behavior based on accumulated statistics, rather than starting from scratch with each new bar.
The article introduces the concepts and tools for working with random processes in trading: definitions, characteristics, autocorrelation functions, and a practical classification. The article covers white noise, random walks, Wiener and Poisson processes, as well as Markov chains and martingales. MQL5 scripts demonstrate the generation of sample paths and allow you to simulate equity, highlighting the mathematical limitations of strategies.
A from‑scratch VMD library for MetaTrader 5: radix‑2 transform, ADMM updates, mirrored windows, and a result type that reports solve quality and mode separation. We verify the solver to full double precision against a reference and test causality by sabotage. The article quantifies repaint from batch versus causal runs and compares it to volatility‑matched random walks. You get a causal indicator, usable diagnostics, and settings that avoid common pitfalls.
This part adds library-wide validation to the double-bottom toolkit. VALIDATE ALL applies the single-sample rule engine to the stored collection, counts its four outcomes and exports a separate CSV report. The report includes settings, sample context and detailed results. Source samples remain unchanged, and export errors are reported separately from completed validation.
This article builds forecast reconciliation in pure MQL5 and applies it to two structures: timeframe aggregation on one symbol and a currency graph linked by no‑arbitrage. The engine projects base forecasts onto linear constraints to make them coherent. We benchmark OLS/WLS and MinT with and without shrinkage on long walk‑forward runs, show when total squared error decreases, and measure the cost that may concentrate in a single node.
We implement surrogate testing for indicators in MQL5: generate synthetic series that match your data's distribution and spectrum, score any statistic, and locate the observation within its null distribution. The library provides four null models, an FFT that handles arbitrary lengths, and a plug‑in interface for custom measures. An indicator visualizes the statistical alongside its chance band for direct interpretation.
The article describes the architecture of a trading system consisting of 20 Grok (xAI)-based LLM agents, each of which embodies a unique trading philosophy — from pure momentum to a statistical z-score. The system applies a genetic algorithm directly during trading: every 20 trades, it automatically eliminates weak agents, clones strong ones with a prompt mutation, and publishes a leaderboard on the MetaTrader 5 chart — without interrupting trading and without using any SDK.
This part adds a visual review layer to the double-bottom workflow. It checks the chart context, moves to the recorded period, and renders the range, two qualifying lows, neckline, classification, score, and the first decisive close. The annotations mirror the validator's indices and prices, helping you audit outcomes without changing sample data.
The framework evaluates boundary sensitivity by rebuilding shifted target-timeframe bars from completed M1 sources and running one Strategy Tester pass for each boundary offset. It includes a ShiftedBarBuilder, a demonstration EA, and a deterministic self-test. An XAUUSD M15 case study reports net profit, Profit Factor, maximum relative equity drawdown, and total trades across all 15 offsets to quantify dependence on the bar grid.
We implement the Harris Hawks Optimization algorithm in MQL5 and analyze five movement modes of agents controlled by a single parameter — the decaying escape energy E. The C_AO_HHO class is introduced; it is compatible with the unified test bench and features a reproducible implementation of Lévy flight. The algorithm was tested on the Hilly, Forest, and Megacity functions with 5, 25, and 500 coordinates — the results indicate anomalous behavior.
The article builds an Ollivier–Ricci curvature library in MQL5 for multi‑symbol analysis. A basket becomes a correlation network, and the reading captures network geometry rather than correlation level. It includes an exact optimal transport solver with an internal certificate, a rolling fragility metric with two informative nulls, an indicator that plots the reading against its null band, and a demonstration expert that uses it as a risk gate.
This work adds an explainable validation module to the recording toolkit for double bottoms. It finds and filters swing lows, selects the neckline, quantifies depth, low similarity, symmetry and recovery, and assigns a rule-based status and scores accepted pairs. The settings are configurable, the process is reproducible, and the original sample data remains intact for comparison across rule sets.
We demonstrate live forecast charting with Kronos in MQL5 and an evaluation pipeline designed to be look‑ahead‑safe. Deterministic dumps are scored per horizon against a random‑walk baseline (15,113 predictions), then fed to a simple Strategy Tester rule on EURUSD H1. The findings show no edge, and illustrate how a high threshold can yield a persuasive but low‑information backtest with very few trades.
This recorder bridges visual labeling and systematic research in MQL5. After you position a rectangle, the EA extracts the candles it spans, derives candle geometry, normalizes OHLC values, assigns an ID, and saves the sample for reuse across sessions. The current category is the double bottom, and the module records the chosen sequence without enforcing a predefined pattern definition.
This article delivers a native MQL5 MiniRocket: 84 fixed convolution kernels yield 9,996 features quickly and deterministically, requiring no training loop and no external runtime. We verify the port against sktime and a float64 reimplementation, then run a reproducible audit of seven classic setups; planted and coin‑flip controls confirm correctness, and a 25%‑flipped control sets the detection threshold.
This article implements the Maximal Information Coefficient (MINE) for MQL5, including the grid search with dynamic programming and the four MINE statistics. It explains why raw MIC has a nonzero noise floor and builds a permutation null to judge significance. The result is a verified library with a dependence scanner and a chart indicator, allowing you to test features and interpret scores consistently across relationship shapes.
This work delivers an MQL5 Expert Advisor that reads open-position symbols, builds return series, and computes a rolling Pearson correlation matrix, displayed as a CCanvas heatmap. Off-diagonal pairs that meet a warning threshold are bordered for quick scanning. The code details return processing, correlation arithmetic, and matrix indexing, with a verification script, helping you track shifting co-movements across your active book.
Learn how to bring advanced plotting and data visualization to the MetaTrader 5 terminal by leveraging Matplotlib, one of Python's most powerful and widely used visualization libraries.
This article explains how to use the Z-Score to quantify price deviations in standard deviations and apply the concept in MQL5. We cover the calculation, interpretation across market regimes, and the implementation of a custom indicator and three trading strategies as Expert Advisors. Readers get complete code examples and a testing workflow, plus practical notes on lookback selection, fat tails, execution timing, and risk controls.
We show that nested cross-validation removes optimistic evaluation bias but does not stop hyperparameter search from overfitting a finite sample, and we add a diagnostic for when the criterion cannot resolve candidates from noise. We implement OffsetPurgedKFold (boundary-shifted repartitioning), a criterion variance audit (resolution ratio and selection stability), a selection optimization curve, a corrected one-standard-error rule, noise-floor early stopping, and an effective trial count for correlated trials. The result is a dispersion report per candidate and an effective count for the Deflated Sharpe Ratio.
This article builds an MQL5 dashboard that evaluates entry timing for closed positions using tick‑based Maximum Adverse Excursion (MAE) and Maximum Favorable Excursion (MFE). It computes an entry efficiency ratio, renders an MAE/MFE scatter plot and an efficiency histogram, and prints a concise summary to the Experts tab. Deal grouping, tick‑level excursion logic, and a standalone verification script are included so you can diagnose entry quality across many trades.
We give our AI chat assistant two new abilities in MQL5: sight and tool access. It can now capture the chart as a screenshot, attach it to a message, and view it in the panel, and it can call tools that read live positions, trade history, indicator values, chart objects, and the economic calendar. The assistant reasons from what it sees and queries rather than text alone.
An on-chart MQL5 indicator generates a live stream of encoded candlestick symbols and a frequency report for the last N bars. It supports color customization and type filtering, and uses A/a, H/h, E/e, G/g, D, and N/n classifications. Traders can quickly review occurrences and common sequences directly on the chart.
This article presents a full MQL5 pipeline for Toeplitz Inverse Covariance Clustering: stacked observations, ADMM‑based graphical lasso with block‑Toeplitz constraints, and dynamic‑programming regime assignment. It separates structure from volatility and conditions away the shared USD leg. The fitted regimes are displayed causally as a non-repainting ribbon with a dependency graph for the active state.
This part of the series examines the mechanisms of the Law of Large Numbers (LLN) and the Central Limit Theorem (CLT) as the theoretical foundation for understanding market patterns. This section describes the tools of descriptive statistics and methods for finding point and interval estimates of distribution parameters. Particular attention is paid to the methodology for testing statistical hypotheses, which makes it possible to objectively distinguish genuine market anomalies from random noise. Each theoretical concept is accompanied by a practical example in the appendix, which helps reinforce the material using specific data.
We implement a Time Series Shapelet library for MQL5 that finds the subsequence of price history whose z-normalized shape best separates two labels and derives the decision threshold from information gain. Because candidate searches on prices always return a winner, the fit includes a purged hold-out and a calibrated block permutation null. You get a reusable facade and rules you can plot and evaluate against an explicit noise floor.
The article presents an MQL5 script that reads closed deals, groups them by date, and renders a calendar grid for daily P&L and trade count. It explains the data model, normalization to midnight, robust week/weekday mapping, diverging and single-hue color scales, and CCanvas-based drawing, plus a verification script and weekday summary. This helps you quickly locate clusters and quiet periods and cross-check findings with logged totals.