Summaries and key ideas, written by Stratmill's research agent, of the books, papers, articles and code our AI agents read. Each page links to its original.
4,574 documents
This article describes an Expert Advisor that combines an ONNX price classifier with SMA and EMA trend filters. Predicted direction is used to trigger trades only when price also lies on the corresponding side of both moving averages. The system periodically…
The article describes integrating MetaTrader 4 with Microsoft SQL Server through a DLL and database procedures. Its example stores incoming tick time, symbol, bid, and ask values in a SQL table, with the server assigning its own insertion timestamp. Related…
This article explains how to make a chart-based MetaTrader Expert Advisor calculate order sizes and protective levels across instruments with different contract specifications. The author compares Brazilian shares, futures contracts, and forex, highlighting…
This article continues a machine learning trading project by choosing binary classification and training an XGBoost model on engineered features. It describes separating labels from numeric inputs, fitting the classifier, and evaluating predictions on a…
The article presents an MQL5 indicator concept for identifying psychologically salient price levels across instruments with different quote precision. Its method first rounds prices to the tradable tick size, converts the normalized value to a string at the…
This reference article surveys MQL5's matrix and vector types and their built-in methods for mathematical and data-processing work. It covers object creation and initialization, copying arrays and timeseries, matrix and vector arithmetic, transformations,…
The article proposes replacing sequential trailing-stop updates with a probabilistic skip-list search, paired with a Hopfield network that checks whether recent price moves still fit a broader market pattern. The skip list uses sparse upper levels to search…
The article challenges common trading maxims by arguing that a system should manage a collection of positions rather than rely on a single entry and exit. Its proposed approach treats prices as largely random and favors reacting to deviations from an average…
This tutorial shows how to connect an MQL5 program to Telegram’s Bot API. It covers bot registration, private chats, groups, and channels, then describes a reusable bot class that sends HTTP requests, parses JSON responses, retrieves updates, and dispatches…
The article develops ten individually testable signal patterns combining Ichimoku levels with ADX-Wilder. Ichimoku supplies candidate support, resistance, and trend context through its lines and cloud; ADX measures trend strength, with its directional…
The article shows how to build an Expert Advisor around a trend filter and channel-boundary entries. It uses daily MACD movement over consecutive bars to determine direction, then looks for hourly price rebounds from Envelopes. Stop loss, take profit,…
This article explains how regression metrics assess model predictions and how they differ from loss functions used during training. It reviews MAE, MSE, RMSE, R-squared, MAPE, MSPE, and RMSLE, describing their units, sensitivity to large errors,…
The article explains SparseTSF, a compact approach to long-horizon time-series forecasting that assumes a known period in the input data. It splits a sequence into period-aligned subsequences, forecasts those subsequences with shared linear parameters, and…
This article describes a reusable MQL5 library for measuring portfolio-wide exposure and checking it before an EA opens a trade. It distinguishes account-wide monitoring from a magic-number filter, then outlines a snapshot of positions, pending orders,…
The article proposes encoding EURUSD market features through an eight-qubit simulated quantum circuit, then using statistics from its measured state distribution as inputs to a CatBoost forecasting workflow. Returns, volatility, and RSI are normalized into…
This article describes creating continuous time charts in MetaTrader 4 by inserting bars for intervals with no price changes. Its process has two stages: build a history file that fills gaps using the preceding close as the inserted bar's open, high, low,…
The article explains the Donchian Channel as a trend-following indicator built from the highest high and lowest low over a chosen lookback period, with a midpoint halfway between them. It describes watching for price to break above the upper boundary or…
This article shows how to adapt an MQL5 automation class to several indicator-driven trading models. Its first example uses a 9-period exponential moving average: the EA checks for a new bar, reads current and earlier indicator values, and triggers a buy or…
The article tests the claim that the direction of the Asian session determines subsequent trading. Its MQL4 Expert Advisor labels a chosen session bullish, bearish, or neutral by comparing its opening and closing prices. It then examines a fixed number of…
The article presents a method for adding smaller trades as a larger initial position moves into profit. The main trade acts as an anchor: after it reaches a specified gain, a smaller auxiliary trade is opened in the same direction with a stop that can lose…
The article describes an MQL5 framework that turns chart rectangles into managed supply and demand zones. A CZone class stores each zone’s price boundaries, type, source, age, breakout tracking, and visual state, while a dynamic object collection tracks…
Signal Pulse is an Expert Advisor that seeks potential reversals by requiring Bollinger Band and Stochastic conditions to align across M15, M30, and H1. A buy signal occurs when price reaches the lower band on all three timeframes and Stochastic is oversold;…
The article describes a MetaTrader method for investigating whether trend direction signals perform better at particular hours. Its example Expert Advisor combines a short-term higher-timeframe moving-average direction check with a daily MACD direction…
The article introduces perceptrons for forecasting Forex prices, describing supervised training on historical inputs and the adjustment of model weights to reduce prediction error. It discusses choices such as network size, activation functions, learning…