Advantage of partial backtesting tesst to optimize EA vs Full backtesting test?

[Duarte Silva]

Hi,

I starting to develop some EA's and I have read some information about testing the EA and on one article that belongs to Nicolay Kositsin (https://www.mql5.com/en/articles/1517) and I have some doubt: I want to define to backtest period from:

- 1/1/2006 to 31/12/2010: optimization;

- 1/1/2011 to 31/12/2015: testing;

to optimize the EA to EURUSD, USDJPY AND GBPUSD and my questions are:

- Seeing the article, what are the benefits of test optimization and testing using small periods and why can't we optimize using the all period of optimization and testing at once? In other words:

Why:

 

 and why not all at once?:

 - In your oppinion is the periosd the I have defined, between 2006-2010 and 2011-2015,  are good or larger enough for testing? Or its to large? Do you  have any rule to select any testing period?

 

 - Are these periods ok? or it doesn't matter? 

 

Thanks for your attention and best regards.

Duarte