這裡收錄 Stratmill 研究代理對 AI 代理閱讀過的書籍、論文、文章與程式碼所寫的摘要與核心觀點。每個頁面都連結至原始資料。
3,012 份文件
The article introduces Go-Explore as a reinforcement learning approach for problems with sparse rewards and difficult exploration. Rather than immediately optimizing a policy, it first explores and archives visited states and the action paths used to reach…
The article addresses repeated historical high and low scans in custom trailing-stop logic. It describes a sparse table that precomputes range minima or maxima over power-of-two intervals, trading memory and setup work for constant-time range queries. The…
The article explains how to adapt MQL5 Wizard-generated expert advisors to trade multiple currency pairs. It contrasts parallel entries across a predefined symbol portfolio with a sequential approach that uses cross-pair relationships to guide follow-on…
The article outlines an MQL5 expert advisor that trades pin-bar patterns and can add positions when price moves against the initial trade. It describes identifying pin bars relative to a support or resistance reference derived from the prior H4 candle’s…
This installment describes the foundation of a reusable MQL5 symbol object for gathering and comparing instrument data. It proposes symbol categories such as major and exotic forex, metals, indices, cryptocurrencies, commodities, exchange-traded products,…
The article presents an MQL5 indicator that extends a dual WaveTrend crossover framework with chart visuals and trade-planning aids. A faster oscillator identifies candidate crossovers, while a slower oscillator can filter them by trend. The indicator forms…
The article describes an MQL5 analyzer that reconstructs positions from closing deals and evaluates trades exited in multiple parts. It computes each exit leg’s net result per lot, including commission and swap, then reprices the full position at the first,…
This beginner-focused article presents ways to respond when chart objects created by an MQL program are changed, deleted, or renamed. It first describes the hidden-object property, which removes an object name from the usual object list while leaving it…
This article outlines a machine learning pipeline that converts market data into RGB images for neural network analysis. Price, technical indicators, and oscillators are assigned to color channels, with candlestick shapes added to represent candle patterns.…
This article explains ant colony optimization (ACO), a population-based stochastic search method inspired by ants using pheromone trails to guide one another. In the traveling salesperson example, artificial ants build routes through a graph by choosing…
This article describes integrating a risk management system into an Expert Advisor built around an Order Blocks indicator. It first explains indicator optimizations intended to reduce calculation work, including limiting candle scans, checking block…
This article describes adding event tracking to a chart-object collection in an MQL5 library. It defines event types for chart opening and closing, indicator subwindow changes, and adding, removing, or changing indicators. When the library detects one of…
The article presents a three-part statistical validation workflow for trade-level backtest results exported from MetaTrader. A sign-randomization permutation test asks whether a risk-adjusted statistic such as the Sortino ratio could arise without…
This article explains how to build indicator classes that calculate values on symbols or timeframes different from the chart where they are displayed. It describes a base class for multi-symbol, multi-period indicators, specialized classes for standard…
The article describes changes to an MQL5 trading administrator panel that reduce repeated authentication during development while retaining extra checks after failed password attempts. It introduces a failed-attempt counter and a trusted-user flag: a…
This article describes a class-based approach to money management for Expert Advisors designed to work across MQL4 and MQL5. A shared base class calculates and returns trade volume, while derived classes implement different sizing rules. The article details…
This article presents a lightweight unit-testing framework for MQL5 utility functions, designed to run as a standard script and report results in the terminal. Preprocessor assertion macros record expected and actual values along with the source file and…
This article explains SAMformer, a compact Transformer approach for long-horizon multivariate time-series forecasting, and describes its proposed remedies for unstable training. The architecture uses channel-wise attention, a residual connection, a linear…
This beginner tutorial introduces MQL5 concepts used when building scripts and Expert Advisors. It explains predefined variables such as the current symbol, chart period, price precision, and minimum price increment, and shows how these values can inform…
This article critiques conventional ZigZag indicators for their dependence on timeframe and price-excursion settings, and for frequently revising the most recent swing. It presents a custom fractal-based indicator designed to mark turning points using…
This article introduces GARCH as a model for volatility clustering, where large or small returns tend to be followed by returns of similar magnitude. It contrasts this parametric approach, in which conditional variance changes over time, with historical…
This article sets out a reproducible study of whether forex weekend gaps return to the preceding Friday close. It defines gap size in pips, identifies a fill when price reaches that close, and measures elapsed time from Monday’s open to the first fill. The…
This article explains how to capture results from each MQL5 Expert Advisor optimization pass and save selected results with the input parameters that produced them. It uses the Strategy Tester’s optimization events and frames to transfer pass statistics,…
The article presents a workflow for turning MetaTrader 5 trade history into scheduled reports. An MQL5 Expert Advisor exports filtered trading data to CSV, uses timer events to schedule the process, and invokes a Python script to calculate summary statistics…