Discussing the strategy VECTORIAL DAX (M5)

Forums ProRealTime English forum ProOrder support Discussing the strategy VECTORIAL DAX (M5)

Viewing 15 posts - 811 through 825 (of 1,239 total)
  • #142858

    where in the code does it calculate the ‘positionsize’ for which PositionSizeLong and PositionSizeShort are derived?

    At line 7 you can have positionsize = any value you like. With MM= 0 it won’t change.

    I have been running a 5m long version of this over the past couple of months. It’s basically a merger between Vectorial and MoD, so far fairly consistent with the back test.

     

    3 users thanked author for this post.
    #142863

    Welcome back nonetheless 🙂

    1 user thanked author for this post.
    #142866

    @nonetheless

    Welcome back 🙂

    I put Your last code above in PRT but my backtest is different from Yours. I also needed to add an “endif” in the end. I’m running version 10.3
    Have I done something wrong?

    #142868

    Hello nonetheless

    Hope you are well.  Many thanks for your reply.  I just tried your suggestion with the code you posted above and it worked.  I will try to incorporate this back into the version of the Vectorial code I had downloaded preciously.

    Thanks

    S

    #142869

    my backtest is different from Yours

    hi Artemois, live forward tests will always vary from the back test for multiple reasons. Also, I think mine was stopped and restarted some time in August so the MM will have reset. But when I run a back test for those 2 months, I see v similar performance ~91%

    and yes, it does need another endif – got lost in the copy/paste

    #142870

    Also, please note that i didnt do any WF or VRT for this; the straight back test looked good and I was curious about the basic concept so I put it on demo for a couple of months, that’s all. I’m sure it can be improved

    #142873

    Hi Nonetheless

    My bad, I didn’t understand it was Your actual result but thought it was the backtest 😉

    Below is a backtest for the same dates.

    #142883

    I got same backtest result as Artemois … different / reduced  by a factor of 15 ish from Nonetheless Pro-Order results.

    Big difference … something must be wrong?

    #142890

    This is well weird … I ran Nonetheless code above again 2 mins ago and got completely different results than I was getting last night!

    Nearer / more comparable with None’s now … see attached.  I still get far more drawdown though … mine dd at £947 vs None’s dd at £64

    What do you get this morning @Artemois ??

     

    #142896

    I’m putting attached on Forward Test on Monday … MM turned off and Period = 55 at Line 37.

    Above is all I explored, I’m sure there would be other beneficial changes?

    It be great if we had a Short version or Long and Short in same System?

    Thank You for sharing Nonetheless!

    #142907

    Hi Grahal

    I get the same results today as yesterday

    I made a backtest on the DJ 3m Vectorial V3 for the same period and it has been more succesfull.
    I also made one of  200 kpts

    I find them both interesting!

    Thank You  @Nonetheless  for sharing, it is very interesting as a beginner to se what the more experienced people are doing.

     

    #142921

    Hi Grahal, how’s it going?

    No idea why you got such varying results from this. I have attached the itf, just in case something got corrupted in the copy/paste. Also the 200k back tests for long and short, with a comparison look at the 3m v3. Obviously the latter is only 13 months, as opposed to 32.

    I also found some tests that I did back in the spring of the 5m long version … seems to check out. But as I said, I really did this out of curiosity to try marrying Balmora’s Vectorial with MoD … and then I kinda forgot about it.

    1 user thanked author for this post.
    #142923

    more attachments.

    MM=0, positionsize = .5

    5 users thanked author for this post.
    #142932

    Both the Vectorial and the MOD are very interesting codes. MOD has a slightly bigger problem with DD but apart from that it has potential. This, and especially the 3 min version is good.
    Is there a chance of any further development ?

    #143034

    Ciao Balmora74 vedo che il tuo sistema anche su timeframe ad 1 minuto funziona bene ottimizzato, tu cosa ne pensi?

    Hi Balmora74 I see that your system even on 1 minute timeframe works well optimized, what do you think?

Viewing 15 posts - 811 through 825 (of 1,239 total)

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