Pathfinder Trading System

Forums ProRealTime English forum ProOrder support Pathfinder Trading System

Tagged: 

Viewing 15 posts - 1,636 through 1,650 (of 1,835 total)
  • #63859

    To trade the PF US500 in ProOrder you have to change line 51 to 5 and line 53/54/55 to 50. The minimum tradesize is 5 according to IG.

    You get more or less the same outcome in the results as with the original sizes. Which is weird because if you take a look at the orderlist in the beginning of 2010

    you see a couple of times orders with 2 positions only. The trades should have been rejected in live by the way.

    And in this code the startrisk is 2. If you take less risk (<2)  you probably have to change the minimumsize as well.

    Regards,

     

     

     

    2 users thanked author for this post.
    #63890

    Hello.

    I’ve been following you here for a longer while. It’s very interesting and I can see how much work is already done in the Pathfinder trading systems.
    Many thanks to all of you for this!

     

    I would like to share my results with you in demo mode.
    Here attached.

     

    And I would like to ask you two three things too.

    Is there anyone who runs the Pathfinder in the Dax live? If so, how long you do this and would you be willing to share the results?

    What results have been obtained over the longest period of time tested with which system?

    And last but not least.
    Does anyone or can someone provide one of the Pathfinder systems as mt4. file? If not, do we want to pool money and hire a professional programming service?

    until then
    JohnScher

    Translated with http://www.DeepL.com/Translator

    #63997

    Pathfinder Silver looks a lot better if you can backtest with 200.000 candles. So I am running it as well.

    I am now live with 13 PF-4H algos. Wow, I hope they perform like in the last 2 weeks. Not a big hit, but all the small profits summed up to a nice chunk of cash.

    I actually also hope they will not trigger all at once, this could severly crush my margin.
    I will definitly reduce the algos when I hit 50%.


    @wp01
    : Thanks for the tip with the Us 500. I changed it accordingly and updated the dropbox.

    1 user thanked author for this post.
    #64183

    life/demo and backtest: backtest needs an additional candle before the data is synchronized again

    From yesterday a position was opened on NIKKEI, but nothing on backtest. And there was more than 4h.

    Is it normal?

    #64322
    Jan

    Thanks a lot for sharing the code of the Pathfinder.  Very organised and very well explained what it is doing in the code.

    #64327

    life/demo and backtest: backtest needs an additional candle before the data is synchronized again

    From yesterday a position was opened on NIKKEI, but nothing on backtest. And there was more than 4h. Is it normal?

     

    Guys, i checked the TS on Nikkei, and it should not have open a trade on 26 of febr, but it did. And i lost 1.200$ for it, could someone help me? @reiner @nicolas

     

     

    #64333

    @Gianluca,

    I loaded PF Nikkei V6 to check the latest positions and V6 also opened a position on the same date and time. I see in your tradelog that the opening was from PF V7, but

    have you checked that you also activated the correct version in ProOrder or that maybe an old version is hanging there? I’m asking because your BT also show two versions.

    Dajvop and Pfeiler are also running PF Nikkei V7. Maybe they can tell you if they had the same positions.

    Kind regards,

    #64335

    Hello Wp thank you for your reply, the version on the screenshot are almost the same, i am running the code that i’ve posted, no trade on Backtest, trade in real :-/ Could you try to backtest it the same?

    #64336

    @Gianluca,

    I already loaded your mac version and i get the same results as you have.

    When did you activate it in ProOrder? There is a date on the right in ProOrder. I’m asking because i had some issues in the past when i activated a code in ProOrder while it is in position,

    live and backtest were not always in sync. So if you activated it between 16-02-2018 09:00 hours and 26-02-2018 05:00 hours you were not have all the buys. I’m not sure if it goes well when the position

    is closed. I can remember i had a few times a problem with that in the past. But this is easy to check for you. If it was before that date than this could not be the problem and i’m out of options.

     

    #64356

    Mmm i started the code on 20 of february…i need to try from that day, i will next week now i will be out for holiday.

    #64402

    @gianluca and @wp01

    I activated the latest NIKKEI v7 on Feb 6. The first buy was on Feb 16 0900, the second on Feb 19 0500 and the third on Feb 19 0900. Now, the backtest and live trading differs.

    BT buys on Feb 21 0900 and then exits on Feb 26 0500.

    LT exits all positions on Feb 21 1700, then buys on Feb 26 2100 and exits with loss on Mar 1 1946.

    #64407

    Thanks dajvop. The latest trade is the same as Gianluca reported after he activated it on the 20th. of february.

    Do you have an idea why it bought on the 26th. at 09.00 and why it didn’t show up in the BT?

    My idea is that it bought a position because of the dynamic positionsize. The algo had a profit in the week before.

    IF DayOfWeek <> DayOfWeek[1] AND DayOfWeek = 1 THEN
    IF StrategyProfit > profitLastWeek + 1 THEN
    positionSize = MIN(trendMultiplier, positionSize + 1) // increase risk
    ELSE
    positionSize = MAX(1, positionSize – 1) // decrease risk
    ENDIF
    profitLastWeek = strategyProfi

    You also had the full profit because you activated it earlier instead of Gianluca. He activated it the 20th. According to your tradelog from above he didn’t had one of the buy orders.

    I think the reason why the results differ from the LT is that when you start the algo, it also starts with 1 position, depending on the risk you take. But when you look at the BT for over

    5 or more years it had already increased multi positions and risk because of the positive results. Looking at the dynamic positionsize it increases the positionsize looking at results and

    seasonal multiplier. (see line 47 and 48). When activated in LT it basicly starts at zero.

    #64553

    @wp01

    You are correct.

    If I start the BT on Feb 6, then the result is identical to LT.

    #65571

    Buongiorno a tutti e un ringraziamento speciale a Reiner per la sua disponibilità, ha preso a disposizione il suo sistema pathfinder. Ho notato che quando uso pathfinder in modo reale le operazioni sono differenti da quelli eseguiti su backtest. Ho anche notato che, in modo reale, le operazioni dipendono dal momento in cui si avvia il sistema. La mia domanda è: qual è il momento giusto si deve lanciare il sistema? Qual è il momento giusto per le operazioni in modo reale coincidere con quelli in modo backtest? Grazie mille per l’eventuale risposta.

    #65572
    reb

    Hello Ant.g

    Il mio italiano non e sufficiente per una risposta

    This post is in the english forum, so could you be kind to continue in this language ?

    thks

    Reb

     

Viewing 15 posts - 1,636 through 1,650 (of 1,835 total)

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