Using Tick Entropy to Adapt Volatility Risk and Trading Signals
Summary
The article describes a MetaTrader 5 and Python system that uses rolling price data to estimate market uncertainty and adjust trading decisions. It calculates Shannon entropy from returns and squared returns, adds volatility and trend features, and feeds the resulting feature set to a neural network that estimates directional probability. A volatility regime detector then informs position size and stop and target distances. An MQL5 Expert Advisor sends recent prices to a Flask server and may execute returned signals subject to exposure and cooldown checks.
The article outlines historical data collection, feature engineering, model training, server inference, and live execution. It presents an implementation architecture and feature calculations, but the supplied text does not report a backtest, out-of-sample results, or evidence that the predictions are profitable. The approach depends on model training and deployment choices, and tick-based requests introduce operational and execution considerations. Treat the claimed rapid adaptation and risk benefits as design goals rather than demonstrated performance.
Key ideas
- Shannon entropy of returns and squared returns is used to represent price uncertainty and volatility uncertainty.
- The feature set combines entropy with return dispersion, higher moments, trend strength, and RSI.
- A neural network estimates directional probability, while a volatility regime detector adjusts trade risk parameters.
- An MQL5 Expert Advisor sends recent prices to a Flask inference service and can act on returned signals.
- The article describes a system design but provides no reported trading performance validation.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.