Hi,
Unfortunately I can not explain why I get different results in the results of Builder and Retester.
Why is that?
Here is an example for audcad:
I received 19 results in the builder. I passed these results to the retester with the same settings. The result of the retester is: 11 of 19 Results are failed! How can that be, the settings are the same and the results differ?
The database for the test is Tickdata to audcad from 2006 to 2019.
Monte Carlo method is random, it will generate different simulations every time you run it. YOu also use a very liittle number of simulations - only 10 or 30.
That way there is a big chance that totally different conditions will be generated every time you run it.
This is why it could have failed in Retester when it passed in Builder.
One solution would be to use much more simulations - at least 100, possibly 500 or 1000 in MOnte Carlo methods. Ths will make it less random and more statistically significant.