Discussing the article: "Neural Networks in Trading: Time Series Forecasting Using Adaptive Modal Decomposition (Final Part)"

 

Check out the new article: Neural Networks in Trading: Time Series Forecasting Using Adaptive Modal Decomposition (Final Part).

The article discusses the adaptation and practical implementation of the ACEFormer framework using MQL5 in the context of algorithmic trading. It presents key architectural decisions, training features, and model testing results on real data.

The testing results are presented below.

Overall, the model generated a positive return during the testing period by executing 13 trades. Slightly more than half of these positions were closed profitably. However, it should be noted that 13 trades over a three-month evaluation period represents relatively low trading activity.

One possible explanation is the nature of the probabilistic attention mechanism itself.


Author: Dmitriy Gizlyk

 
Are the 13 trades for a single currency pair? If 10 pairs are included in the analysis, how many trades will there be?