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 implements temporal features in MQL5 to match a Python feature pipeline for machine-learning models. It covers Fourier encodings for hour, weekday, and day of year; flags for major forex sessions and their overlap; session-conditioned rolling…
The article translates the Darvas box method into an MQL5 Expert Advisor. It defines a box from a price extreme that holds for three sessions, then requires a breakout above the box boundary with increased volume. The implementation adds filters for low…
The article describes a universal MetaTrader 4 and 5 template that runs robots for multiple symbols from one chart. It explains the motivation: simplify setup, reduce the load of many chart instances, and evaluate a combined multi-instrument equity curve.…
The article describes TrendMap, a signal-generation framework that combines swing-based Fibonacci retracement levels with VWAP. It calculates a selected bar range’s high and low, derives retracement levels, and compares VWAP with the midpoint of that range:…
The article explains how legacy and updated MQL4 tools create and update offline charts from standard chart data. Its example uses a period-conversion script on EURUSD one-minute bars to generate custom two-, three-, or four-minute history, which is then…
The article describes a MetaTrader 5 expert advisor designed to manage trades across multiple currency pairs and two metals from one chart. Its entry logic combines a seven-period ADX directional movement signal with Parabolic SAR confirmation. The ADX rules…
The article compares an EURUSD hourly LSTM model trained on unfiltered prices with versions trained after low-pass filtering, then explores training on February data concatenated across years. It also varies the filter settings and prediction horizon across…
The article walks through preparing financial time-series data and training a small language model on a CPU. It introduces tokenizer roles and distinctions between encoder, decoder, and encoder-decoder models, then describes collecting price data from…
The article outlines a MetaTrader 5 Expert Advisor that accepts a configurable list of currency pairs and uses one designated pair as the signal source for trading the others. Its proposed relationship logic distinguishes pairs that share a selected base or…
The article develops modifications to a damped linear growth exponential smoothing model for financial time series. First, it changes parameter fitting from minimizing one-step forecast error to minimizing the combined squared errors for one-, two-, and…
The article explains how to fine-tune a pre-trained GPT-2 model with adapter modules and compare this approach with LoRA and full-parameter fine-tuning. Its adapter design projects hidden features into a smaller bottleneck, applies a nonlinear activation and…
The article introduces the Stochastic Oscillator as a measure of where a closing price sits within the high-low range over a selected lookback period. It explains the faster %K line and smoothed %D line, then describes fast and slow variants and walks…
The article explains how trading activity shifts across the Asian, European, and American sessions and how those shifts can shape intraday price movement, liquidity, and volatility. It introduces the ISI ProSpread SMA indicator as a way to study hourly…
This article describes an MQL5 Economic Calendar filter that restricts new entries around scheduled events. It loads and caches calendar data, filters events by importance, and compares each event’s currency with currencies associated with the traded symbol.…
This article explains how to draw rounded speech bubbles for MQL5 trading interfaces by combining a rounded rectangle with a triangular pointer. The pointer can face up, down, left, or right, and its position can be shifted relative to the bubble body. The…
This article presents an MQL5 Expert Advisor that places interactive labels beside open positions on a chart. The goal is to connect each position with current price action and show position details and profit or loss without relying only on the toolbox…
This article shows how an MQL5 Expert Advisor can monitor account activity through the OnTradeTransaction event handler. A trade operation may generate several notifications as a request is processed, an order enters or leaves the active list, and deals and…
This article explains why recurrent networks may suit market time series: they carry a hidden state forward, allowing information from earlier observations to influence later processing without repeatedly supplying a large fixed input window. It describes an…
The article explains how discriminant analysis can classify market observations into groups such as upward and downward price moves. It illustrates the idea with technical indicators: statistical analysis can identify variables that contribute little to…
This article describes an MQL5 Expert Advisor that turns swing highs, swing lows, and pullback logic into multi-timeframe reversal trades. It tracks recent and prior swing points on higher and lower timeframes, assigns bullish, bearish, or neutral structure…
The article evaluates whether the CBOE Euro and Gold volatility indexes (EVZ and GVZ) add forecasting value for XAUEUR. It describes merging the indexes with daily XAUEUR OHLC data, imputing missing values with column means, and examining correlations,…
The article describes an automated forex system that builds a matrix of exchange rates for eight major currencies and estimates pairwise fair values from direct quotes and cross rates. It uses currency transitivity—for example, deriving one cross rate…
The article examines how trading activity varies across intraday sessions and argues that fixed-time bars can make indicator observations statistically uneven. It proposes operational time, measured by a fixed number of ticks per bar, as a way to normalize…
The article addresses weaknesses in randomly sampled training labels for a market classifier, including class imbalance, serial dependence among examples, and overlap between classes. It uses principal component analysis to visualize feature structure and…