Thanks for the article!
Ran a TestEMD Validate on the TestEMD EA, suggesting trading a month after optimising for the last year (InSampleDays = 365, OutSampleDays = 30), and got these pictures on D1 for 01.01.2018-27.03.2020.
EURUSD:
GBPUSD:
AUDUSD:
USDCHF:
USDJPY:
In general, only EURUSD showed more or less stable results.

- www.mql5.com
Thank you very much for this interesting article but the files do not work, they have syntax errors and undeclared variables.
Regards
Thank you very much for this interesting article but the files don't work, they have syntax flaws and undeclared variables.
regards
Please, provide your error logs and specify exactly what you did.
Thanks to Stanislav for the very good article. I downloaded the codes and meet an error while compiling it.
The error reads: 'Offset' -some operator expected in the line
buffers[0].set(Offset,sum);
Please show how to solve that.
Thanks
Thanks to Stanislav for the very good article. I downloaded the codes and meet an error while compiling it.
The error reads: 'Offset' -some operator expected in the line
buffers[0].set(Offset,sum);
Please show how to solve that.
Thanks
I have just downloaded the sources anew and compiled all indicators and the expert without an issue.
You're doing something wrong. Make sure you have extracted the contents of the archive with preserved folder structure.

- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
New article Forecasting Time Series (Part 1): Empirical Mode Decomposition (EMD) Method has been published:
This article deals with the theory and practical use of the algorithm for forecasting time series, based on the empirical decomposition mode. It proposes the MQL implementation of this method and presents test indicators and Expert Advisors.
A single test with these settings for the period from early 2018 up to February 2020, i.e., with a forward for the year 2019 and early 2020, paints the following picture:
TestEMD report on EURUSD D1, 2018-2020
As we can see, the system benefits, although the indices show that there is a room for improvement. Particularly, it is logical to assume that more frequent reoptimization in a step-by-step mode and search for the step size can improve the performance of the robot.
Basically, it can be said that the EMD algorithm allows identifying on larger timeframes the fundamental, in some sense momentum fluctuations of quotes and create a profitable trading system based thereon.
Author: Stanislav Korotky