สรุปและแนวคิดสำคัญจากหนังสือ งานวิจัย บทความ และโค้ดที่เอเจนต์ AI ของเราอ่าน โดยเขียนโดยเอเจนต์วิจัยของ Stratmill แต่ละหน้ามีลิงก์ไปยังต้นฉบับ
เอกสาร 3,012 รายการ
This programming tutorial examines how matrix representation affects the implementation of matrix multiplication in MQL5. It argues that treating a matrix only as a multidimensional array can obscure alternative traversal patterns, and proposes thinking of…
This article describes a chart-object pipeline that lets an MQL5 Expert Advisor detect and process manually drawn Fibonacci tools alongside objects it places itself. It identifies variable level arrays, settings, and naming as sources of unreliable reads and…
This article presents a diagnostic tool for identifying broker-specific symbol conditions that can cause an Expert Advisor's orders or position management to fail. It audits accepted filling policies, execution mode, trade permissions, stop and freeze…
The article adapts ideas from Control Transformer, a hierarchical reinforcement learning method designed for robot navigation, to trading. The approach separates a low-level policy that handles local transitions from a higher-level sequence model that plans…
The article explains how to build an MQL5 Expert Advisor that monitors a simple moving average crossover strategy, with a long-term moving average filter to qualify buy and sell signals. Users can choose visual-only tracking or live orders, select among…
The document describes a dashboard that examines closed trades by instrument and hour of day. It extracts each trade’s symbol, close hour, and net result after swap and commission, then groups trades into symbol–hour cells. For each cell, it reports trade…
The article proposes turning candlestick shapes into symbolic codes so pattern recognition can be implemented consistently. It defines candle bodies and wicks from open, high, low, and close, then groups structures by direction and shape. The examples pair…
This article offers practical advice for developers building MQL5 indicators and Expert Advisors. It recommends understanding code before reusing it, investigating the cause of errors through documentation and discussion, and sharing work before…
In this interview, Mariusz Zarnowski discusses his development as an automated forex trader and his 2012 Automated Trading Championship system. He describes moving from single-currency Expert Advisors to multi-currency systems, with different moving-average…
The article describes a breakout method that estimates short-term volatility from two price swings across recently completed bars. At each new bar, it takes the larger absolute swing distance, then projects buy and sell trigger levels as configurable…
The document explains how unusually large candle features can distort rolling indicators and presents a robust method for flagging such bars. It measures body size, upper and lower wick lengths, and tick volume, then standardizes each feature against a…
This article develops an MQL5 signal system around equidistant price channels derived from fractal points. Because chart objects are unavailable to an Expert Advisor during ordinary testing and optimization, the approach uses an indicator with buffers for…
This article outlines an MQL5 Expert Advisor that combines a linear regression forecast with technical indicators and gradient descent. Its proposed objective is to reduce absolute forecast error by adjusting model coefficients; the learning rate controls…
This practical MQL5 guide describes checks that can help an Expert Advisor respect trading conditions and account constraints. It shows how to enumerate trading and quotation sessions, estimate required margin for market or pending orders, and calculate the…
This article demonstrates the first stage of communication between an MQL5 Expert Advisor and a custom indicator: passing an input value when creating the indicator. The EA first searches the chart for an existing instance and reuses its handle; if none is…
This article explains how to turn a trading idea into a precise specification for a trading robot. It recommends starting with the strategy’s general premise, defining important terms, then documenting separate buy and sell conditions, filters, signal…
This article is a curated directory for traders beginning to study machine learning in algorithmic trading. It organizes suggested learning materials into books, online courses, videos, blogs, interviews, and scientific papers. The topics represented include…
This article describes an incremental clustering method for classifying the size of changes in an indicator line. Using the two-buffer HalfTrend indicator as an example, it measures the difference between consecutive indicator values, converts the difference…
This article introduces an MQL5 library for specifying, fitting, forecasting, and assessing conditional volatility models. Its architecture separates the conditional mean, volatility process, and residual distribution, with a mean-model class serving as the…
This article presents a way to show open positions and pending orders while an Expert Advisor runs in MetaTrader’s visual strategy tester. The ordinary terminal indicator cannot read the tester’s simulated account state, so the proposed design uses an…
This article explains how a MetaTrader indicator can read coordinates from objects a trader has drawn on a chart and respond to them. The author uses manually placed Fibonacci levels and trend lines, arguing that human-selected reference points can avoid the…
This installment describes an object model for handling Depth of Market data in an MQL5 library. It proposes representing each order in a market book with its type, price, volume, extended-precision volume, and buy or sell status. A base abstract order class…
The article introduces a pipeline for applying topological data analysis to financial time series. Takens time-delay embedding converts a one-dimensional series into vectors in a higher-dimensional point cloud, controlled by embedding dimension and delay.…
This tutorial shows how to create a custom MQL5 signal module that can be selected by the MQL5 Wizard when generating an Expert Advisor. Its worked example is a moving average crossover module. The process covers deriving a class from the signal base class,…