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 explains how to construct a price channel in MQL4 from confirmed fractal turning points. It searches a timeframe-specific bar range for two upper or lower extrema, projects a line through them, then locates an opposing extremum to draw a parallel…
The article extends an MQL5 ARIMA class with easier prediction methods and model saving and loading. It explains why forecasts need enough input observations: autoregressive terms require lagged values, non-contiguous lags preserve intervening time…
The article explains an MQL5 method for proposing Elliott Wave labels from price-chart turning points. It first outlines classical impulse and corrective-wave structures and their constraints, then describes two broad approaches to automatic labeling:…
In this 2011 interview, Sergey Nikitin describes the Expert Advisor that was leading the Automated Trading Championship during its second week. It traded EURUSD and EURJPY, used daily charts to determine trend, and relied on trend-watching indicators.…
This article shows how to replace hand-written candlestick rules with TA-Lib pattern functions in a Python service connected to a MetaTrader 5 Expert Advisor. The EA sends recent OHLC data to a Flask endpoint; Python builds a time-indexed DataFrame,…
This article presents a channel-breakout entry method intended to capture a change in trend. Channels can slope upward, slope downward, or move sideways, and may be drawn using several channel techniques. The proposed signal occurs when a candle closes…
This article describes a MetaTrader 5 expert advisor that can trade multiple forex and metal pairs from one chart, with an option to run on a single pair. Its signal design combines ZigZag swing structure with the Awesome Oscillator (AO), allowing several…
The article applies supervised learning to three Parabolic SAR and Relative Vigor Index signal patterns that previously failed forward-walk tests. It describes implementing the indicators in Python, connecting to MetaTrader price data, and training a neural…
The article introduces Markov chains as probabilistic models for transitions among defined states. It explains the transition matrix and the memoryless assumption: the next state depends on the current state and elapsed time, rather than the full path taken…
This article describes how to extend an MQL5 economic calendar dashboard with controls for currency, event importance, and time range. Users can toggle the filters from the chart panel, select currencies, and use a cancel control to clear filters and remove…
The article explores ways to combine the Fractal Adaptive Moving Average (FrAMA) with the Force Index in an MQL5 trading system. FrAMA is used to describe trend direction or slope, while the Force Index contributes volume-weighted momentum information. The…
This article explains a workflow for deploying a trained Torch time-series forecasting model in a MetaTrader 5 Expert Advisor using ONNX. It covers switching the model into inference mode, collecting and naming its inputs and outputs, exporting it, then…
This article presents ways to make MetaTrader 5 Expert Advisor optimization more selective. Instead of enumerating every available timeframe, the author proposes exposing a curated set of timeframes as an input parameter, so early searches compare broad…
Sergey Pankratyev describes the design of his Automated Trading Championship robot, which trades twelve currency pairs through a single complex EA. It uses technical analysis and market reaction data, with parameters adjusted from trade entry and exit…
This article turns the DiNapoli approach to Fibonacci retracements into an automated strategy. It uses retracement levels to define possible entries and expansion levels to set profit objectives, while moving averages filter trades by trend and Stochastic…
The article investigates whether planetary positions, lunar phases, solar activity, and planetary aspects can help forecast EUR/USD movements. It describes collecting daily astronomical data, aligning it with market history, and applying correlation analysis…
This article argues that financial machine learning performance depends heavily on how the prediction target is defined. Unlike a fixed medical label, a trading target may represent returns, price appreciation, volatility, drawdown, or relative asset…
The article develops an automated USDJPY strategy on daily bars using bullish and bearish engulfing candles as reversal signals. It defines the patterns by comparing the current candle’s open and close with the prior day’s prices, then uses the average true…
The article describes martingale approaches that add to losing positions, either by increasing trade size or scaling into a position at improving prices. The aim is for a later favorable move or pullback to recover earlier losses. It suggests that strict…
The article proposes using mutual information (MI) to choose which of several candidate trading targets a black-box regression model should learn. The examples compare future changes in price, moving averages, Stochastic, and RSI. MI is presented as…
The article outlines a multi-symbol expert advisor that tracks a two-hour price range for each selected instrument, from 10:00 to 12:00 UTC+2. After the range closes, an upside break can trigger a buy when the Stochastic reading is at or below 20; a downside…
The article outlines a multi-symbol Expert Advisor that evaluates mean-reversion and momentum signals separately for each instrument. It calculates a Z-score from price relative to a moving average and standard deviation, and measures momentum as a price…
The article describes a MetaTrader Expert Advisor for finding and marking common single, double, and triple candlestick patterns. Users can search for one named formation or scan all supported patterns over a selected historical window. A chart panel…
The article explains how the Goertzel algorithm computes selected discrete Fourier transform frequencies efficiently, making it an option when only a limited range of cycle periods is of interest. It describes a class that scans periods, returns real and…