wp01Participant
Master
Miguel,
Can you please show me a picture of the results after you make the changes like you said (window 8-23) and avoiding reopening intraday day in the same direction?
Attached the results from my backtest from february 2010. Still an amazing factor of 3,77.
Thanks.
Regards,
Patrick
Fr7Participant
Master
Hola Reiner,muchas gracias por tu gran aporte.Tengo dos peticiones:
1)Ya puesto,sólo te falta para el SP500,podrías crear el sistema…….
2)Puedes crear un sistema con un timeframe de 1 hora.
GRACIAS
@Fr7
Please speak English in the English forums. You can use online translation if you can’t. Thanks.
Hi Reiner,
Any plans of converting your great code to the volatile Hang Seng index?
Best regards,
David
however, it would be advisable not to waste energy and time on many indexes, better to focus on a few and improve them.
Hola Fr7,
Fr7 asked for a S&P version in 1H timeframe.
Pathfinder framework will work more or less on every index. Some work better and others are less suitable. Requirement for a suitable index is volatility. I have tested a S&P Pathfinder version but the results were positiv not comparable to DAX or FTSE due less S&P volatility.
Pathfinder is working in the moment only for 4H. I didn’t test other timeframes so far, may be later. In the moment I want focus only on H4.
David, Hang Seng and Nikkei are on my watchlist. In the moment I’m working on commodities and I’ m very keen to have something for oil, gold and silver.
Brent crude oil is very volatile.
I Can try to check this out:)
And post as soon as its done.
Thanks for all your energy and time to comit to such a work as you have done.
Hi everyone here, I’d like to join in to help Reiner improve his strategy like many others, but I do not have enough time unfortunately actually. But it seems very interesting and I hope everyone here appreciate the spirit of sharing and caring of Reiner 🙂
Well done guys, go on like this, automated trading is a long and rough journey, but hopefully great at the end!
Hi Kenneth, I appreciate every support for this project. Go ahead and share your results.
Nicolas your support can be is of great help to the group and to the system. I hope you can find time for this cause. thank you.
Hi all,
I’ve been following the PathFinder project for some time, and it seems promising. I’ve been running v2 live since a few months and it has been good so far but want to get over to v4 preferably.
But now I have an issue I can’t really figure out. Right now on my live account with v4 it has taken a long position. But running the same code (exactly, copied it from the live system), the backtest does not take this position. I have not previously seen any differences between live/backtest with PathFinder, but I have mostly been running v2.
In the attached image, to the left is the live system, to the right is the backtest of v2, v3 and v4. I’ve tried changing the start date of the v4 backtest to be the same as the live system as this should be the only thing that would be different, but it does not seem to matter either.
The code running in v4 is the same standard DAX v4 as earlier in this thread, with only some modifications to the PositionSize.
I can’t really figure out that there could be anything that could cause this? Can anyone else see the same thing or find an explanation? (I might surely be missing something..)
Falmskarm, PRT needs one additional candle to synchronize real trade events. The long entry was on 1700 so backtest will be synchron at 2100
Thanks Reiner, a bit too eager it seems 🙂 I checked it and it’s just as you say.. The thought crossed my mind but I could not understand why it would need that logically. I mostly work with shorter timeframes except with PathFinder so I haven’t really noticed. Great job with PathFinder!
