intraday DAX strategy 5min mini1€ spread 1

Forums ProRealTime English forum ProOrder support intraday DAX strategy 5min mini1€ spread 1

Viewing 15 posts - 346 through 360 (of 503 total)
  • #23385
    CN

    Paris, I don’t understand?

    #23386

    just a clue  : these are backtest from 27 january to today …… you ve got it now  ?

    #23388

    I totally agree !!!!

    #23389

     

    all of us  know the bactkest is positive and winning in long term .

    now we had some losses last days , but we had losses in backtest too in the same period …..

    what i think it is not bit trouble .

     

     

    #23390
    CN

    Paris, still dont get it. Please explain

     

    #23392

    backtest wins

     

    if  real/live losses and backtest losses  , it means  it is normal to have these last losses from 27 january  .

    even if  all of us see that backtest is not exactly the real/live .

     

     

    #23395

    I think it’s difficult to create a ”scalp” system with tight stops and profits. Because of the spread and slippage. We build the system based on backtests. Backtests have no slippage. If you have a large stop and profit, the slippage doesn’t matter. But with systems like this with tight trailling stops it’s hard to make it profitable because backtests aren’t so reliable.

    #23400

     

    @ raul and nicolas

    @ all

     

    For the TS issue  :

    in my opinion , what we have to do is replicate the trailling stop of the backtest in real , i mean to code exactly what it does in backtest and applicate it in real . And not to use the ” native ”  TS  of  pro real time .

     

    But for to do this we have perhaps to apply it in a tick for tick graph , not a 5 min . which mean all code have to be recode to be efficient in a tick for tick graph .

     

    for example if in the code  we have 18 candles of 5 minutes we must convert on 18 x 5 minutes on new code .and so on

     

    what do you think ?

     

     

    #23401

    Hello, today a negative day again, let’s see how it progresses. The main problem is the shifts in price. I understand that you can not run a robot code with the graph in ticks mode. Would there be some way to target the operation in 1 minute graphics? That is, if at 9.05 the price is higher than at 9.00, the order ends. Of purchase, the same as until now but in graphs of minute.

    #23405

    raul said : I understand that you can not run a robot code with the graph in ticks mode.

    why ? i disagree . If you adapte the code it could be nearly the same that in 5 minutes .in all cases better the system in live .

    #23406

    I mean pro real time does not let a robot activate a temporality tick by tick, try to put the graphs in tick by tick and give you to boot any robot with this temporality. It is not a problem to adapt the code, it is a problem of pro real time. At least the last time i tried i did not leave

    #23408

     

    @ raul

    and don t you have another way to code it ?

    could gave us  some samples of code that will not work on tick by tick ?

    and why not re code the system for 1 min and keep the temporality of 5 min , for one candle of 5 min we will have 5 candle . Maybe is enough to be efficient . we don t have to be exactly the same , if the difference is some points is ok .

     

     

     

    #23411

    @ raul sorry for all these questions .

    what disapointed me is that before th 10.3 versions,  they said that the issues and problems will be resolved with new version , with the  new tick by tick backtest.

    Now  we ve got it there are some issues again .

    #23427

    hi, another loss again on Dax v3 in my real account, PRT v10.2

    #23438

    Hi Nicolas,

    Apologies for the delay, it took me some time to put together the detail you requested. Yes the strategy utilises pending stop orders. So please see attached the PRT executed screenshot of the relevant trades. I have then put together in excel every single action in the account for these trades (if you would like the excel sheet itself please let me know, I wasn’t sure if you are allowed to upload here). I have carefully found and grouped together each trade’s action activity in time order and in alternate colour for each set of trades so it is easier on the eye to follow. Hope this is what you require. Let me know if you need anything else.

    With regards to your other question, yes I agree with your conclusion (given the exceptions of spread and slippage etc that we know about) that the issue seems to lie with how the trailing stops are implemented and executed. The position opening orders seem to work ok within tolerance limits of Live vs Backtest. This is also proven by the fact that if you look at my previous screenshot, the big issue is usually always with the closing – ie. trailing stop trades rather than the opening ones.

    Many thanks for looking into it.

Viewing 15 posts - 346 through 360 (of 503 total)

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