Kokkuvõtted ja põhiideed raamatutest, teadustöödest, artiklitest ja koodist, mida meie AI-agendid loevad. Need on koostanud Stratmilli uurimisagent. Igal lehel on link originaalile.
3,012 dokumenti
This article introduces Transformer architecture and focuses on the Temporal Fusion Transformer (TFT) for multi-horizon time-series forecasting. It describes attention, parallel training, positional information, and differences among encoder, decoder, and…
This article presents an engineering framework that turns manually drawn MetaTrader 5 trendlines into managed objects with identity, interaction history, and lifecycle state. A manager discovers and registers chart lines, while separate components handle…
This article describes a multi-symbol Expert Advisor that checks market execution conditions before acting on a liquidity-sweep continuation signal. Its filters monitor spread expansion, tick velocity, quote gaps, short-term volatility, slippage estimates,…
This article adapts an online analytical processing library for quote data as well as trading records. It separates shared cube components from trade-specific selectors and records, then adds tools for grouping observations by month or weekday, calculating…
This article explains how to extend an MQL5 library so an indicator can retrieve and display price data for selected symbols and timeframes. It reorganizes shared data definitions, adds arrays for chosen symbols and periods, and describes a separate…
This article presents the later patterns in a series combining the Gator oscillator, which reflects trend activity, with the Accumulation/Distribution (A/D) line as a volume-based check. Pattern 5 looks for breakouts that lack A/D confirmation and describes…
The article presents an MQL5 indicator intended to automate the drawing and classification of support and resistance. Its method scans a configurable number of candles for extreme turning points, then looks for additional tests near those prices, using a…
The article describes an MQL5 indicator that ranks support and resistance zones by volume participation within the currently visible chart area. It detects the viewport’s time and price boundaries, identifies candidate levels using configurable wick- or…
The article argues that a model’s forecasts may be difficult to compare directly with observed prices because predictions and targets can occupy differently transformed coordinate spaces. It explains this idea through a geometric interpretation of linear…
The article describes adding divergence signals to a trend-constrained trading expert advisor. Regular bullish divergence pairs lower price lows with higher indicator lows, while regular bearish divergence pairs higher price highs with lower indicator highs.…
The article examines how uneven bullish and bearish labels can bias financial classification models toward the more common class. It illustrates the issue with daily candle direction counts across several instruments and a USDJPY Random Forest example.…
The article explains Symbolic Aggregate Approximation (SAX) for converting price windows into compact symbolic sequences. It first normalizes each window to emphasize shape, averages values within equal segments, then maps those averages to symbols using…
Royal Flush Optimization (RFO) adapts genetic algorithms by representing each solution coordinate as a discrete sector rank rather than a binary string. It initializes a population of candidate solutions, selects stronger candidates, exchanges coordinate…
This installment develops the visual controls for an MQL5 Expert Advisor’s order system. It adds a bitmap close or cancel button to chart objects that represent trade levels, positions the button alongside those objects, and expands an edit-field class to…
This article develops an MQL5 indicator that uses market-depth volume to identify candidate bullish and bearish order blocks, then exposes signals and plotted take-profit and stop-loss levels for automated trading. The proposed pattern rules combine a volume…
This article explains how MQL5 represents the lifecycle of a trade and how a history library can collect and organize completed trading activity. It distinguishes orders, deals, active positions, and closed positions, describing how identifiers and history…
This article presents a practical implementation of DUET’s channel clustering stage for multivariate time-series forecasting. It transforms each channel with a Fourier method, compares frequency-amplitude vectors, normalizes the pairwise distances, and uses…
The article develops ten rule-based patterns combining the DeMarker oscillator with Envelope bands. Examples include fading a move that crosses a band and returns inside when DeMarker is extreme, following consecutive closes outside a band when momentum…
The article outlines a computer vision workflow for classifying EURUSD price direction from OHLC time series. It retrieves hourly data from MetaTrader 5, forms rolling windows, scales each window, and labels it according to whether the closing price is…
The article turns the Oops gap reversal pattern into an MQL5 Expert Advisor. A bullish setup begins when a bar opens below the prior bar’s low by at least a configured gap threshold; a bearish setup mirrors this above the prior high. The EA stores the gap…
The article tests whether sequential bootstrapping reduces correlation among bagged trees trained on overlapping triple-barrier labels. It separates two changes often made together: reducing the number of sampled rows to reflect average uniqueness, and…
This article explains how correlation between price and time can quantify the direction and strength of a market move. It introduces Pearson correlation on raw values, Spearman correlation on ranks, Kendall’s rank measure, and Fechner’s sign-based measure. A…
This article presents the practical architecture and training of StockFormer, a hybrid trading framework combining predictive coding with reinforcement learning. Its three predictive branches model relationships among assets and forecast market behavior at…
This article describes an automated process for periodically optimizing a MetaTrader 4 Expert Advisor and updating its live parameters. A separate tester terminal runs an optimization over recent historical data, writes a report, and the main terminal…