Question on genetic optimisation - page 10

 
Angela писал(а) >>

Can you draw any conclusions based on this number? I can't run it on a larger interval, I don't have a minute history. The TS is not optimized, I gave up on optimization, nothing good comes out of it, it just wastes a lot of time. I have adjusted my entries/exits using August's history (trades from 82 to 93). However I don't know what to do further on, even 10% drawdown is bothering me, and in this case it's more than 14%, if MM is used, it may increase many times over.

In my opinion this is the minimum required number of deals and then I have to "master and master" it means to look for possibilities to improve indexes.

for example, one of my ways

after optimization, choose results with more or less decent indicators, but certainly with a large number of deals, for example, 200-300

then connect different filters (e.g., temporal) and get better results, while the number of deals will decrease (they will be filtered), then switch to demo mode, and so on.

everyone has their own ways and their own Expert Advisors, so mine might not suit you.

 

Can you tell me if there is a limit to the number of records when saving a file in the batter?

I was collecting statistics in visualization mode for 2 years, and after I finished, I saved the file, the number of records is the same as the number of calculated bars, but at the end of the array instead of data are all zeros.

The size of the array is 145687*117, I got zeros starting from 139172 records, although the processing was going on in the tester and the charts were built to the end.