Discussing the article: "Neural Networks in Trading: Integrating Chaos Theory into Time Series Forecasting (Final Part)"

[MetaQuotes]

Check out the new article: Neural Networks in Trading: Integrating Chaos Theory into Time Series Forecasting (Final Part).

We continue to integrate methods proposed by the authors of the Attraos framework into trading models. Let me remind you that this framework uses concepts of chaos theory to solve time series forecasting problems, interpreting them as projections of multidimensional chaotic dynamic systems.

Testing used data from January–February 2025. This period was chosen to ensure a rigorous evaluation on previously unseen data. All other experimental parameters remained unchanged to ensure reproducibility and a fair comparison. This methodology eliminates random factors and allows objective assessment of algorithm performance.

The testing results are presented below.

 

During testing, the model executed 287 trades, with nearly 39% closed profitably. Despite the relatively low win rate, the strategy produced a positive overall result due to the profit-to-loss ratio. Specifically, the average profit per winning trade was twice the average loss, compensating for less successful trades and yielding an overall positive outcome, with a profit factor of 1.15.

The average position holding time exceeded 2 hours, indicating a tendency for short- and medium-term decisions. Notably, the longest-held position lasted nearly two days. This fact requires further analysis.

Author: Dmitriy Gizlyk