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
The article proposes evaluating a complex trading system by measuring its component blocks separately with shared criteria, rather than optimizing every parameter at once. It uses net profit, maximum drawdown, and the share of profitable trades to compare…
The article describes how to represent an aggregate position formed from multiple orders in MetaTrader 5's hedging account mode. It classifies positions as buy-only, sell-only, net-buy, net-sell, or locked, with mixed positions distinguished by comparing…
The article compares two forecasting targets for a Bollinger Bands trading approach: whether price will rise or fall over a future horizon, and which of four zones around the bands price will occupy. It describes labeling historical GBPUSD daily data,…
AutoGraf is a MetaTrader 4 application that uses chart objects and a control panel to manage orders graphically. Traders can create market and pending orders by dragging lines, and set or change stop loss and take profit levels through corresponding chart…
The article presents a framework for a MetaTrader 5 Expert Advisor that processes multiple currency symbols while maintaining separate settings for each. It compares three ways to schedule checks: timer events, ticks on the chart symbol, and chart events…
This article explains trend scanning as an alternative to assigning every market observation a fixed prediction horizon. For each starting point, it fits linear trends over a range of forward windows, calculates a t statistic for each, and selects the window…
The article presents a proposed three part model of past, current, and future transient functions, labeled P, C, and F. It says the functions are derived from modifications of the Gamma distribution and satisfy a normalization relationship in which their sum…
This article describes a price action Expert Advisor that identifies swing highs and lows, labels successive swings as higher highs, lower highs, higher lows, or lower lows, and uses those sequences to infer trend direction. A bullish Change of Character…
The article introduces machine learning for price prediction, then outlines a workflow using MetaTrader 5 hourly EUR/USD data, Pandas, scikit-learn, and neural networks. It explains basic decision trees and the GRU architecture, comparing its reset and…
The article proposes using neural networks to filter signals from an adaptive version of Thomas DeMark's Sequential strategy. It focuses on Setup and Intersection signals near potential turning points, treating the network as a decision aid rather than a…
The article develops a grid-like martingale EA that opens additional positions at fixed adverse price intervals, increasing trade size along the chain. It closes the whole basket when aggregate profit reaches a configured target. Since the approach can…
The article introduces artificial neurons and multilayer networks, describing inputs, weights, weighted sums, activation functions, and how learning changes weights. It explains scaling inputs to a bounded range and compares step, sigmoid, and hyperbolic…
The article examines a reverse martingale approach in which a losing trade is followed by an opposite-direction trade with increased size, aiming for a later win to recover prior losses. It discusses practical settings based on the author’s tests:…
The article describes a VWAP-based signal tool that connects a MetaTrader Expert Advisor with a Python analysis service. The EA gathers recent candle prices and volume, checks for invalid values, saves the data, and sends it for processing. Python libraries…
This article describes an MQL5 Expert Advisor that monitors multiple forex and metal symbols from one chart and generates signals when a fast RSI crosses a slow RSI. It explains symbol selection modes, including trading the chart symbol alone or choosing…
This article describes an Expert Advisor that helps traders search a large symbol universe by filtering instruments and simplifying chart navigation. Its configurable filters can exclude symbols based on Market Watch presence, open positions, spread, price…
This article lays out the design and MQL5 implementation of a chart-based panel for manual Forex trading. Its proposed controls include buy and sell actions, position-closing buttons, editable stop-loss and take-profit settings, and displays of account and…
The article outlines a proposed trading system that combines a quantum-circuit state encoder, a deep Q-network, language-model analysis, and a self-adaptation mechanism. It motivates the design with familiar reinforcement-learning difficulties in markets:…
The article proposes measuring whether candles opening at the same time of day repeatedly move in the same direction, and whether changes in that pattern coincide with shifts in trend. Its MQL4 script groups candles by opening hour and minute over a selected…
This article introduces spans, experiments, and compositions from category theory, then uses them as a framework for describing a trailing-stop experiment. It records the floating profit of a long position alongside moving-average and average-true-range…
The article describes an MQL5 script for preparing MetaTrader 5 historical quotes for use in external analysis software. Its worked example targets two versions of NeuroShell DayTrader Professional, which expect different CSV headers and date-time layouts.…
The article introduces simple exponential smoothing and related Holt and Brown models as methods for short-term extrapolation of time series. It explains the central assumption that past and present behavior can inform near-term forecasts, while emphasizing…
The article introduces the discrete Hartley transform (DHT) as a real-valued method for moving a discrete signal between time and frequency representations. It applies DHT to price series, where individual harmonics can be removed, attenuated, selected by…
The article explains OPPO, an offline reinforcement learning approach that learns a policy directly from comparisons between trajectories. It contrasts this with the common two-stage method of first fitting a scalar reward model and then training a policy…