Optimiser – Positions of Variables affects Result!

Forums ProRealTime English forum ProRealTime platform support Optimiser – Positions of Variables affects Result!

Viewing 15 posts - 16 through 30 (of 35 total)
  • #194885

    CAN SOMEONE PLEASE SOLVE THIS ?!?

    This is undoable as the results become totally unpredictable of it. It is life-threatening, so to speak.
    Delete one Optimization Parameter which is not the bottom one and nothing works any more. And this really was not so, two weeks ago.

    #194890

    I sent below direct to PRT on their Contact Form.

    Please check out the link below for an Issue that is giving us problems using the optimiser.

    https://www.prorealcode.com/topic/optimiser-positions-of-variables-affects-result/#post-194536

     

    1 user thanked author for this post.
    #194980

    PRT say we need to send in a Technical Report.

    Do you want to send a Tech Report Peter?

    Does a Tech Report via your IB Account get to PRT any faster than via IG Account??

    Or better still do you have a direct from PRT Account / Platform?? Surely a Tech Report submitted from such a PRT Premium Platform would go direct to PRT??

    #194981

    As far as I know Nicolas is investigating. He will report asap.

     

    1 user thanked author for this post.
    #194989

    Does a Tech Report via your IB Account get to PRT any faster than via IG Account??

    All what goes out from my (or on behalf of my) PRT-IB account is at least always read and responded to (I dare say 100%). But like I told 18 months or so ago, it is a bit difficult to report bugs from a platform which is not only *not* used by me for backtesting, but which also can’t even be used for Autotrading. This conflicts.

     

     

    As far as I know Nicolas is investigating. He will report asap.

    For this moment I really hope that Nicolas can achieve something.

    #195261

    I received below from Victor at PRT re the problem discussed in this Topic.

    We have identified the problem you reported and our technical team is currently working on a correction, which will be implemented as soon as possible.

    Big Thank You from All of Us to PRT!

    1 user thanked author for this post.
    #195341

    Just had that weirdness where I got £6478 Gain in the optimiser Table. Then when optimiser had finished, the  maximum Gain was only £4346 in the optimiser Table!!

    I’m glad I saved the variable values for the £6478 Gain!

    Maybe the bug above is connected / related to the ‘position of variables bug’ discussed in this Topic!?

    #195344

    I saw that day before yesterday again too. But then I imagined that the larger profit which showed at first was a negative amount and was thus later superseded with the positive amounts ? (the negative amount dropping to the bottom of the list ?)
    Nah …

    #195345

    How do you optimize your systems? After maximum profit or minimum drawdown? Personally, I have never encountered this problem, even in shorter periods of time. I always optimize for minimal drawdown.

    #195373

    Maximum profit here. But maybe I can learn something ?
    I never understood the other selections as no matter what, the sorting remains on Profit. Of course I can change the sorting so it sorts on DD (or others) but this is not what the setting seems to be for (attachment 2).

    So how would I accomplish “optimizing on DD” ?

    #195376

    the sorting remains on Profit

    Same for me no matter which selection I make.

    I assumed (wrongly?) that the selections only work on the Premium version??

    After maximum profit or minimum drawdown?

    Max profit for me.

    You would only notice the weirdness (max Gain disappear!) if you are looking at the results as they populate the Table. As I only ever (nearly) optimise with 100 combinations, I mostly watch the Table getting populated.

    I have seen the same weirdness many weeks back.

    #195378

    The order of the variables that impact the results and especially the empty results (0€), has been identified, a fix will be deployed asap.

    1 user thanked author for this post.
    #195407

    Well, I don’t want to know what the maximum I could win, I want to know what the maximum I can lose. That’s why I use minimum drawdown sorting.
    I have no idea if my thinking is correct. But it feels right.

    #195408

    But when we’re at it anyway (apparently not what phoentsz is talking about), maybe @Nicolas knows what this should accomplish ?

    #195412

    It seems to work as it says… but an expert explanation would still be good. 😉

Viewing 15 posts - 16 through 30 (of 35 total)

Create your free account now and post your request to benefit from the help of the community
Register or Login