Discussion re Auto-System – HLHB Trend Catcher – DAX mtf

Forums ProRealTime English forum ProOrder support Discussion re Auto-System – HLHB Trend Catcher – DAX mtf

Viewing 15 posts - 16 through 30 (of 65 total)
  • #105624

    Variables at lines 16-19, 184-186, 190 and 246-251.

    Namely all those that may help the performance of the strategy to keep at par with the evolving market.

    #105626
    AE

    Yes, I see a lot of variables. Do you use walk forward with all of them?? Dam! It will spend a lot of time. Could you please tell us the range that you used per variable?

    Thanks

    #105629

    It takes time to get things working, nothing comes for free.

    I usually test them once in a fortnight or in a month at most.

    You cannot test all variables at one time.

    If a variable has a value of 10 now, I will test it in the range +-2, that is 8-12. If the result is on one of those boundaries, I will test the new likely value in the range +-2 until it does stand in the middle 3 possible values  of the range.

    #105630

    But you will soon realize that most of the variables, if you update them consistently, rarely change.

    1 user thanked author for this post.
    #105902

    Hi Roberto,

    Thanks for coding this – it looks a very interesting strategy. I have it running in demo on the Germany 30 in IG.

    I understand that to initiate a long trade, it requires the 5 EMA to cross over the 10 EMA, the 8 RSI to cross over the 50 mid-line and the candle to be Bullish, BigBullish or HugeBullish on the 1 Hour time frame.

    I note that these conditions were met this morning for a few bars commencing at 8 am. Can you help me to understand that when the ‘event’ “Crosses Over” is true, it means this has happened “for the first time” on the same candle, and is not just simply one measure is above another (5EMA > 10EMA and RSI(8) > Mid(50) )?

    If its the former, then that means the 5EMA must cross over at exactly the same candle as the RSI crosses over the mid-line. Which is clearly a less frequent event than this happening on adjacent or different candles.

    Appreciate your help

    N

     

    #105906

    At lines 214 and 224 conditions are connected with AND, thus needing to be ALL true at the same time (same candlestick, the one at whose closing time the strategy is executed).

    It’s impossible that such conditions last more than ONE bar, once a crossover occurs the next bar it can keep going up or cross the other way round. This is valid for both MA’s and RSI.

     

    #105912
    AE

    But you will soon realize that most of the variables, if you update them consistently, rarely change.

    Thanks Roberto!

    What are the variables that you should change with more frequency? Rigth now, what are the variables that you are using?

    Thanks again.

    #105913

    I am using and checking ALL variables, as you keep running it you will realize which one usually change.

    Currently I am running it on my demo account and never changed one, because demo just hast to tell me if it profitable or not, so I do not need to squeeze the most out of it. Should I run it on my real account next year, then I’ll check setting every other week.

    #105920

    Thanks Roberto.

    So had I wanted to trigger a trade this morning (see attached image) the code would need to read:

    5EMA > 10EMA AND RSI(8) > Mid(50)

    rather than…

    5EMA Crosses Over 10EMA and RSI(8) Crosses Over Mid(50)

    Is this correct?

     

     

     

    #105922

    That’s correct, since the two events occurred on different candles.

    But…. in your case the two events are likely to still be valid for N consecutive candles, thus you might enter several LONG trades with the same conditions if a previous trade closes shortly (maybe within the same bar). It can be risky, because after 4-5 bars a pullback may ensue and your LONG trade could be triggered exactly at that moment with a chance to hit your SL.

    One solution would be to set a flag whenever you enter a trade and clear it when a crossing the other way round occurs.

    1 user thanked author for this post.
    #105929

    Roberto

    I tested your code at 200k and got the result as the picture shows.

    Looks like a stable algo although the result is flat from Feb to June.

    Nice DD :))

    Tick ​​by Tick.

    spread = 1

    3 users thanked author for this post.
    #108473

    System is often stopped in demo and live. An idea of the reason?

    #108487

    What is the reject reason shown in the Rejects Tab in the Orders List?

    I’ll check my version tomorrow and let you know how mine is doing.

    Off to bed now! 🙂

    1 user thanked author for this post.
    #108505

    That’s it. A problem with the others running systems?

     

    #108507

    Another stop of the system today at 10h05 🙁

Viewing 15 posts - 16 through 30 (of 65 total)

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