Pathfinder swing TS

Forums ProRealTime English forum ProOrder support Pathfinder swing TS

Viewing 15 posts - 1,111 through 1,125 (of 2,005 total)
  • #39094

    Palladium and Platinium

    2 users thanked author for this post.
    #39099

    Soybeans

    2 users thanked author for this post.
    #39104

    Thank You O-jay8.

    Really appreciated.

    #39110

    @bluetime6, if you mean a version for the full year, then I don’t think that has been done yet. So far we’re at the first half of July.


    @Patrick
    and @O-jay8, have anybody of you had a chance to look at DAX for July1? I think it might be good.

    I am still trying to get my probacktest to work again…

    #39113

    Hey Dajvop,

    I did a try on Dax and its not that bad but the old version without the Maxprofit.

    1 user thanked author for this post.
    #39118

    DAX:

    regards,

    Patrick

    1 user thanked author for this post.
    #39129

    Problem solved at last. Uninstalled Firefox, Java and PRT, cleaned cache, cleaned registry, booted and reinstalled all, then it worked again 🙂


    @Patrick
    you should’t need to restart PRT after only 2-3 optimisations.

    #39156

    @Oskar Bergval,

    There is a more detailed discription of the optimization posted in a Pdf. file by Mark on page 6. (#20396)

    https://www.prorealcode.com/topic/pathfinder-swing-ts/page/6/

    You say that your optimizations are always worse than the ones posted. Maybe you can post one of yours.

    The only differences we get is that we choose another parameter from the optimization report. It is really no rocket science and i know nothing more than you do.

    Patrick

    #39157

    @dajvop.

    Great job David. In the end it seems a lot of work and also requires more computer skills to get it all work properly again.

    #39232

    Patrick: I used that one. I did one a few pages back which sucked, but I will try more when able. What I meant about a more detailed info is what to change and in which order (after the fixed variables). But maybe it’s just changing the numbers until you get the best result.

    #39269

    My NQ and OMX:

     

    1 user thanked author for this post.
    #39276

    @Patrick, what is the difference with this new OMX and your other one?

    #39278

    @Oskar,

     

    Maybe you can try with one of my files to work back and try to get to the same results. For instance the DAX Jul1 from a couple of post before.

    Import it and optimize periodThirdMA from 1-10 and periodLongMA from 1-100 with an interval of 1. You can try that with the rest of the settings i have and see where it comes with

    and after that, you try the optimization but than with stoplossLong set to 6 and take profitLong set to 4. The Maxcandleeslongwithprofit and without profit you set to the average of 10.

    Once you optimized the first two try to figure out why i took 4 and 87. When you optimize the first two you have to look if you can find one with the highest winning trades. I’ve noticed

    that it is more difficult when the winning trades are about 65% to get to 75% or higher percentage when you optimize the second two (stoploss/takeprofitlong). If you can find a 85% and higher

    of winning trades with the first two optimizations, it is in most cases going to get fine with the rest of the variables. The stoploss and take profit you have to optimize with an interval of 0.25 or 0.1. With 0.1 it

    can take a while if it is 30 years of course, that’s why i take 0.25 interval. In most cases the differences are not that much.

    Further very basic: not more than two variables in one go, set the period to 1. Start with

    accumulation to zero. Actually when you change that to 1 you have to start over again with the rest. You get weird results when you do not have your spread set properly. Always check the spread. And if

    your startingcapital is not set properly you could also get weird results. For example the capital of 10.000 Zar and 100.000 Zar gives you quite different results. The same with OMX: 10.000 or 100.000.

    For exchanges you can check Oanda:  https://www.oanda.com/currency/converter/

    And have you checked if you have the same history in the charts as we post? You can not compare results with different history.

    If you want to create a new optimization you can better duplicate the code from the former month. You go to the part indicaters/tradingsystems, stand on the code from former month and than click duplicate. You rename the month and than you start changing the settings. The advantage if duplication the former one is that everything is already set right and do not have to be change anymore. For example: capital, spread, minimum amount of contracts, names etc. after the // and so on. All these parts are already correct.

    Good luck.

    Patrick

    #39281

    @David,

    Difference in Risk/Reward.

    As attached.

    1 user thanked author for this post.
    #39298

    @Patrick, I meant if there was any difference with the version you posted on page 73. The results seem to be the same to me.

Viewing 15 posts - 1,111 through 1,125 (of 2,005 total)

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