Library / algorithmic

Algorithmic trading

Automation, backtesting and systematic-trading concepts.

• What Is Algorithmic Trading?Rule-based execution — from simple indicator crosses to ML — removes hesitation but not model risk.
• MetaTrader 5 Ecosystem for DevelopersMQL5, Strategy Tester, hedging accounts, and multi-asset support — architecture differs materially from MT4.
• Writing Your First MQL5 Expert AdvisorScaffold OnInit/OnDeinit/OnTick, wrap trading in guards, log everything — smallest viable strategy first.
• High-Quality Tick Data for BacktestingGarbage in → garbage out: stitch ticks from reputable vendors; watch timezone and dividend adjustments for stocks.
• Running Robust Tests in MT5 Strategy TesterChoose modelling quality, control latency assumptions, and validate with forward passes — not a single in-sample run.
• Walk-Forward OptimisationTrain parameters on in-sample windows, validate on subsequent out-of-sample slices — fights curve fitting.
• Avoiding Curve Fitting & Over-OptimisationToo many free parameters + short history = beautiful backtest, fragile live.
• Dynamic Lot & Risk Algorithms Inside EAsScale risk with equity curves, volatility (ATR), and event filters — never static martingale.
• VPS Hosting for 24/7 EA UptimeCo-locate near broker/exchange matching engine; monitor latency and clock drift.
• Martingale & Grid Systems — Risk AnalysisDoubling schemes balloon tail risk; grids assume mean reversion that may never come before margin call.
• Integrating Technical Indicators in MQL5 EAsUse handles, buffer slots, and error checks — recalculate only when new bars form unless tick scalping.
• Time & News Filters to Pause EAsEncode economic calendar IDs, spread thresholds, and session blackouts — killswitches beat hope.
• Monte Carlo Stress Tests for Trading SystemsShuffle trade order, perturb returns, bootstrap paths to estimate drawdown distributions.
• Designing Multi-Currency / Multi-Symbol EAsLoop symbols with correlation-aware risk budgets; watch margin in netting vs hedging accounts.
• Automating Breakout StrategiesDefine objective range/break detectors, false-break filters (retest, time-stop), and volatility gates.
• Minimising Latency in Algorithmic ExecutionMeasure ping vs broker gateway; co-locate; avoid Python bridges for ultra-low latency unless necessary.
• EA Logic Tailored to XAUUSD (Spot Gold)Gold: session-driven volatility, spread spikes, and macro event sensitivity — test separately from FX majors.
• Integrating Machine Learning with Algo TradingFeatures, labels, leakage control, and live drift monitoring — ML is operations-heavy, not plug-and-play.
• Running Multiple EAs as an Algorithmic PortfolioCentral risk ledger, correlation assumptions, and kill switches across experts — avoid double leverage.
• OrderSend() vs OrderSendAsync(): Asynchronous Order Flow in MQL5Blocking vs non-blocking trade requests in MetaTrader 5: latency, OnTradeTransaction reconciliation, and design patterns for multi-symbol EAs.
• SQLite in MQL5: Designing a Local Tick & Feature LedgerSchema design, journaling, WAL trade-offs and crash-safe writes for EA-side research artefacts — offline-first quant workflows on MT5.
• Walk-Forward Matrix: Statistical Design for Honest Out-of-Sample EvaluationRolling train/test grids, multiplicity control, causal features and why a matrix of splits beats single-window storytelling.
• DLL Integration in Expert Advisors: Engineering Risk & ControlWhere native code helps (SIMD, ONNX runtimes), where it hurts (ABI drift, trust boundary), and broker policy constraints.
• Maximum Adverse Excursion (MAE) as an Algorithmic Objective & Diagnostics ToolPath-dependent drawdown inside trades, optimisation pitfalls, linkage to stop placement and expectancy decomposition.
• ONNX Inference from MQL5: Bridging Research ML Models into MetaTraderGraph export, operator support limits, latency budgets, numeric precision and deployment packaging for EAs.
• Event-Driven News EA Architecture on MetaTrader 5OnTimer vs OnTick, calendar APIs, debounced state machines, slippage-aware flattening and kill-switch patterns.
• Multi-Currency Backtesting Architecture in MT5 Strategy TesterModelling margin in deposit currency, cross-rate consistency, portfolio-level metrics and parallel optimisation discipline.
• Latency Engineering for Signal Copy / Copy Trading Distribution FabricsCo-location trade-offs, fan-out buffering, jitter budgets, SLA math between master equity curve and follower fills.
• Stationarity Testing in Financial Time Series — ADF vs KPSS and Null Hypothesis AsymmetryUnit-root null vs stationarity null, size/power trade-offs, structural breaks & rolling windows for live models.
• Volatility Clustering & GARCH-Class Models for Algorithmic Signal ConditioningLeverage asymmetry, fat tails, EWMA shortcuts vs likelihood-based GARCH, and online calibration for prod systems.
• Hidden Markov Models (HMM) for Latent Market Regime Detection in Systematic TradingState-space intuition, EM training pitfalls, feature selection, and mapping latent states to risk & leverage policies.
• Hurst Exponent & Fractal Structure of Price Series — Interpretation GuardrailsH≈0.5 random walk vs trending/mean-reverting mythology, estimation variance, and why Hurst ≠ free alpha.
• HFT Market-Making Algorithms — Skeleton Control Loop & EconomicsQuote revise cycles, skew from inventory toxicity, rebates/fees interplay & fairness debates on speed asymmetry.

Educational glossary content only — not investment advice.