Discussing the article: "Developing a multi-currency Expert Advisor (Part 7): Selecting a group based on forward period"
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Check out the new article: Developing a multi-currency Expert Advisor (Part 7): Selecting a group based on forward period.
Previously, we evaluated the selection of a group of trading strategy instances, with the aim of improving the results of their joint operation, only on the same time period, in which the optimization of individual instances was carried out. Let's see what happens in the forward period.
The standard MetaTrader 5 tester can perform single passes and optimization taking into account the presence of the so-called forward period. When using it, the tester will split the entire specified test period into two parts — main period and forward period. At the end of the main period, all trades are closed and the balance of the trading account is returned to its initial state. Then the EA will work again on the forward period, and all statistics collected by the tester will be calculated separately for the main and forward periods.
In the field of machine learning, the terms In-Sample and Out-Of-Sample (IS and OOS) are often used to refer to the dataset models are trained and tested on. In our area, the main period will play the role of IS, and the forward periodwill be OOS. Keep in mind that OOS is a broader concept than the forward period. We can run EA tests without specifying the forward period in the tester, but if the test period is located outside the period the optimization was carried out on, then this will still be testing on OOS.
Since we have not used OOS testing much in the past (except for the end of this article), then it is time to see whether it is possible to maintain comparable results when working on the forward period, as shown by the EAs on the main period.
Author: Yuriy Bykov