Mother of Dragons trading strategy…

Forums ProRealTime English forum ProOrder support Mother of Dragons trading strategy…

Viewing 15 posts - 391 through 405 (of 523 total)
  • #148620

    The backtests seem fine to me but could be an issue with positionsize. Minimum for demo is 1, minimum live is .5

    #148633

    But didn’t obtain same variables as you

    Probably because mine is optimized with 200k of data. But also, I don’t necessarily choose variables for max profit, sometimes I think it’s better to go for a higher %win esp if it gives higher number of trades. Sometimes it’s based on VRT results.

    As for adapting to other instruments, you really just have to go through all the variables for each timeframe top to bottom. You’ll have to do this 2 or 3 times as one change effects another. I have tried versions for DAX, CAC, SP500, FTSE, ASX – cant remember what else. Some of them are still limping along in demo but the DJ and NAS are the only ones I have faith in. If anything else might work, I would start with the DAX and try to bang that into shape, using DJ v5 as a template.

    #148640

    @nonetheless  It is the same time as in Spain that comes in “NAS 5m MoD v4S” On November 1 the time changes in the US and then it will be necessary to put if not onmarket or (time <= 153000 – USDLS and time> = 220000 – USDLS) and it keeps giving me the same failure with the MM activated and I have 1 lotti in demo. Doesn’t it give you an error?

     

    #148643

    It tests fine for me, both with or without MM. If the problem only occurs with MM you could try using a higher factor, like 5 0r 10, or try it with volpiemanuele’s MM

    1 user thanked author for this post.
    #148651

    You are right, putting a factor of 5 does not give an error. Could you put what result gives you in 200k of candles with factor 3, please? thanks

    #148653

    I really wouldn’t worry about it, any factor below 10 will give a result in the millions – it’s only something to dream about. But if factor 3 blows up over the past 100k then that can also happen over the coming 100k. I will change mine to 5; if the algo succeeds then positionsize will increase dramatically enough.

    1 user thanked author for this post.
    #148740

    @volpiemanuele @nonetheless (ps. still searching for you)

    Can you think of any reason why MM v5 and V.5a would be giving me these blank results. I am literally using the code that you both posted.

    DJI MOD v5 works perfectly (see attached). It’s really strange.

    This is DJI v5a + MM

    This is DJI MOD v5a code that I am using:

     

    #148744

    @nonetheless – I just realized that in v5a you have MM switched on for some reason?

    That’s what seems to be breaking it for me. Is there any variables in MM that would be tied to the instrument or currency? 

    As soon as I switch MM off it works again, which would also explain why the MM version is not working.

     

    1 user thanked author for this post.
    #148745

    @StingeRe My code run ok in my PRT (demo or live version). Try to change MM inserting this code. If MM = 0 you have fixed position and if MM = 1 you have the MM. I see that your currency is AUD. You can insert the correct value in AUD in this variable “ddpp” and “capital”. My currency is EURO. Thanks

    #148746

    @SitngeRe…first change the value in AUD….for me the problem is this….the code is correct….

    #148766

    I just realized that in v5a you have MM switched on for some reason?

    MM is entirely optional, for me it tests correctly both with or without. If yours wont run with MM=1 you can try altering the positionsize, the factor or both.

    #148783

    @nonetheless, I have some issues with the TS, it just adjusts to BE but does not follow the price forward to cover the open positions. Do you have any idea why? Thanks in advance!

    #148785

    Yeah, I just noticed that as well, I’ll look into it.

    1 user thanked author for this post.
    #148835

    Hopefully this fixes the TS problem.  NAS to follow…

    7 users thanked author for this post.
    #148850

    Thanks @nonetheless, I noticed this last night as well when I was forward testing 4.xx vs 5.

    Interesting they both ended up taking similar profits in the end though.

    Thanks kindly for the fix.

Viewing 15 posts - 391 through 405 (of 523 total)

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