Articles on machine learning in trading

The RAFT framework offers a fundamentally different approach to forecasting market dynamics — not as a one-time snapshot, but as an iterative refinement of the state in real time. It takes both local and global changes into account, maintaining a high degree of accuracy even with complex price structures.

The article implements Online Portfolio Selection in MQL5: exact simplex projection, four allocators (EG, Universal Portfolio, PAMR, OLMAR), and a BCRP solver on a concave objective. We verify weight paths against the published Python reference and self-check the optimizer. A drift-aware transaction cost model measures turnover and net wealth on a multi-asset basket. Readers get a tested library and evidence on when regret-bounded methods lose to costs.

The article continues the work on implementing the STE-FlowNet framework, which combines multi-stream processing with recurrent structures for accurate analysis of complex data. The tests conducted confirmed its stability and flexibility in various scenarios. This architecture speeds up computations and enables deeper modeling of dependencies in time series. This approach opens up new possibilities for practical application in trading and analytics.

The article addresses online false discovery rate control for streaming signals, where fixed thresholds inflate noise discoveries. We port LORD3, LORD++ and SAFFRON into MQL5, construct a causal bootstrap p‑value source without a resolution floor, and verify correctness before market data. Dependence stress tests, a budget/level indicator and a two-arm Strategy Tester show how to configure, monitor and throttle streams to keep the acted-on noise fraction within target.

Permutation tests that assume exchangeable observations miscalibrate feature screens for overlapping financial labels, yielding a tenfold false‑positive inflation on EURUSD. This article implements a calibrated three‑gate pipeline: ADF+KPSS for stationarity, uniform‑grid entropy with modal occupancy, and mutual information corrected via a joint permutation null and Westfall–Young max‑T. The ScreeningReport summarizes gate outcomes and effective test count so you can filter features reliably before fitting models.

An LLM agent with SQL memory in MetaTrader 5 — an architecture that eliminates the main limitation of traditional LLM Expert Advisors (EAs): the loss of context between queries. Each agent decision is recorded in SQLite along with the market conditions at the time of the signal — RSI, MA alignment, volatility — and after the position is closed, the outcome is written to the database. When the next request comes in, the agent reviews its own history — similar past RSI conditions and the last three agent decisions — and only then analyzes the current market. The result is a system that remembers its errors between restarts and adapts its behavior based on accumulated statistics, rather than starting from scratch with each new bar.

We implement the Harris Hawks Optimization algorithm in MQL5 and analyze five movement modes of agents controlled by a single parameter — the decaying escape energy E. The C_AO_HHO class is introduced; it is compatible with the unified test bench and features a reproducible implementation of Lévy flight. The algorithm was tested on the Hilly, Forest, and Megacity functions with 5, 25, and 500 coordinates — the results indicate anomalous behavior.

The article builds an Ollivier–Ricci curvature library in MQL5 for multi‑symbol analysis. A basket becomes a correlation network, and the reading captures network geometry rather than correlation level. It includes an exact optimal transport solver with an internal certificate, a rolling fragility metric with two informative nulls, an indicator that plots the reading against its null band, and a demonstration expert that uses it as a risk gate.