Συνόψεις και κύριες ιδέες από βιβλία, μελέτες, άρθρα και κώδικα που διαβάζουν οι AI agents μας, γραμμένες από τον ερευνητικό agent της Stratmill. Κάθε σελίδα παραπέμπει στο πρωτότυπο.
3,012 έγγραφα
The article outlines an MQL5 system for classifying market behavior as trending, ranging, or volatile. It motivates regime awareness by explaining that strategies built for one condition may struggle when the market changes. The proposed foundation combines…
The article presents an MQL5 architecture that turns AI analysis into structured signals an Expert Advisor can route, parse, display, and potentially execute. A central dispatcher maps stable action identifiers to separate workflows, each with its own data…
This tutorial introduces the MetaTrader 5 Economic Calendar and explains how its event data can be accessed from MQL5 programs. Calendar entries include country and currency associations, event descriptions, importance, scheduled times, and reported values…
This article describes chart indicators for inspecting MetaTrader 5 Strategy Tester runs. A reusable chart template can load a selected indicator automatically during visual testing. The Ticks indicator displays recent bid and ask ticks, shows their timing,…
The article explains how neural networks can support trading systems, focusing on signal generation and indicator calculation. It distinguishes forecasting a future closing price from producing a decision signal during a trading session, and argues that the…
This article describes a multilayer perceptron that forecasts a conditional Gaussian distribution for price increments rather than producing only a point estimate. It motivates the approach by noting that mean squared error corresponds to a Gaussian…
This article presents ways to combine classifiers whose outputs may be discrete class choices, confidence values, probabilities, or ordinal rankings. It explains why rankings can help compare models with differently scaled outputs and why they are more…
This programming tutorial explains how MQL5 represents date and time values and applies them in trading-system logic. It introduces the datetime type as a count of seconds, shows how to compare and adjust timestamps, and describes conversions between strings…
The article presents StockFormer, a hybrid framework that combines Transformer-based predictive coding with an Actor–Critic reinforcement-learning agent. Three branches learn representations for long-horizon trends, next-day returns, and changing…
This article describes a MetaTrader 4 workflow for collecting profitable Expert Advisor optimization runs across multiple backtests into HTML tables that can be analyzed in spreadsheet software. It proposes recording separate results for long and short…
The article explains how to calculate four common moving averages from price history in MQL5: simple, exponential, smoothed, and linearly weighted. It outlines each method’s weighting behavior and provides indicator implementation examples, starting with a…
This article develops a pending-order strategy based on unusually high tick volume during a candle. It compares current volume with the average from a chosen number of closed candles, then scales the threshold according to a base deviation and an additional…
This article describes an Expert Advisor that combines Bill Williams fractals with 14-period and 200-period exponential moving averages to mark potential breakout entries. For a bullish signal, price must cross above a recent fractal low while trading above…
This article extends a MetaTrader utility for filtering and navigating symbols with separate tabs for all candidates and user-selected long, short, and range lists. It describes storing each tab’s symbols in an array, drawing tab buttons in a loop, showing…
The article presents an indicator that selects two swing points using DeMark-inspired rules, then draws rising or falling trend lines. Candidate peaks and troughs are identified by comparing each bar with its neighbors and applying additional close-price…
The article develops a modified ZigZag indicator and a class intended to make its swing data easier to retrieve and study. Its key adjustment is to use maximum bid prices for highs and minimum ask prices for lows, incorporating spread into the plotted…
The article surveys visual tools in the MetaTrader 5 strategy tester for examining both individual tests and optimization runs. Tester statistics include trading activity by month, day, or hour, distributions of maximum favorable and adverse excursion, and…
The article describes an Expert Advisor that periodically retunes its strategy parameters against a rolling span of historical data. A manager requests candidate parameter sets from an optimization algorithm, runs each set through a virtual copy of the…
This installment explains how a MetaTrader trading engine can coordinate multiple independent strategy instances within one Expert Advisor. Its CStrategyList manager distributes events such as ticks, timers, and book updates to each strategy, while handling…
This article recasts a binary recurrence matrix as the adjacency matrix of an undirected network: observations become nodes, and recurrent pairs become edges, with self-links removed. It explains how network measures such as degree, clustering, transitivity,…
The article describes an Actor–Director–Critic (ADC) deep reinforcement learning design intended to guide policy learning when a conventional Actor–Critic's value estimates are unreliable early in training. The Actor proposes actions, the Critic estimates…
This article proposes transforming technical indicators into features that may be more informative for a perceptron than raw prices or indicator levels. It defines three approaches: distances between indicators or between current and historical values,…
The article discusses account-model differences that matter when adding automation to a MetaTrader Expert Advisor. Netting accounts combine trades in a symbol into one position with an average price, while hedging accounts keep positions separate and can…
This installment describes organizing MQL5 code for projects that use a shared Expert Advisor library. It recommends keeping each project in its own repository and placing each dependency in an Include folder as a separate repository clone. A shared library…