State Space
Forums › ProRealTime English forum › ProOrder support › State Space
- This topic has 17 replies, 5 voices, and was last updated 8 years ago by Nicolas.
-
-
05/20/2016 at 11:28 AM #7476
The objective of this thread is to discuss trading system State Space.
I found it in Renato Di Lorenzo’s book: Trading Systems – Theory and Immediate Practice.
I have mailed Renato Di Lorenzo and its ok to discuss it public here.
I don’t think Renato Di Lorenzo is the author of the system.The system is made for daily timeframe. Because PRTs lack of “look inside bar” its not possible to test and optimize Stop loss, Take profit and Trailing stop.
Therefore the first goal is to get the system to work as well on 5min timeframe as on daily timeframe.Please use all systems here carefully. I have not used them live yet.
I attach post the original system here, and a modification I have made which only open 1 order at a time.
The system do need to be optimized from time to time.!!! Please use all systems here carefully. I have not used them live yet !!!
05/20/2016 at 11:30 AM #7481@Nicolas could you please help to get the system to work as well on 5min timeframe as on daily timeframe?
I have tried a few things but cant get it right.05/20/2016 at 11:32 AM #7483Second goal would be to find out a good exit strategy and money management which take down risk.
05/20/2016 at 11:36 AM #7484Third goal would be to combine this system with the grind system discussed here:
http://www.prorealcode.com/topic/grid-orders-with-one-combined-stop-loss-and-limit-can-it-be-done/State space could then maybe be used as an entry point.
I have tried some combinations which I will attach here.
My favorite combinations so far is the attached. Next post will contain combinations with BB exit.05/20/2016 at 11:38 AM #7490Next two combinations contain the BB exit from the grind system.
I think here that the triggered floating stop loss and spread eats up to much profit.05/20/2016 at 11:40 AM #7495Here is some pictures which may help 1/2
05/20/2016 at 11:41 AM #7501Here is some pictures which may help 2/2
Here we can se how profit is eaten up.05/20/2016 at 11:43 AM #7505Seems like the forum changes names on the .itf files when I upload, I don’t know why?
05/20/2016 at 12:23 PM #7513Great Job Simon!
Very interesting!! It will take a bit to test it for me at the moment but I’ll be glad to ear news about it.
Cheers
1 user thanked author for this post.
05/20/2016 at 10:50 PM #7588Hi Simon!
Interesting, especially if it can be combined with the grid order code to a complete strategy.
Thanks for sharing:)
1 user thanked author for this post.
05/29/2016 at 2:58 PM #8318Could you please see if its possibile to get the system to work as well on 5min timeframe as on daily timeframe?
I have tried a few things but cant get it right.Thanks! All the best!
Simon05/29/2016 at 5:59 PM #8325Hi Simon, sorry I totally missed this topic. Great job 👌
I’ll do my best to help you this week hopefully.
In the meantime, 5 minutes timeframe is full of noise and a lot of strategies can’t be adapted easily because of that. Anyway I’ll have a look. Got some bugs fixes to be done for the site and improvements as well this week, so I’ll do my best. 😉1 user thanked author for this post.
05/30/2016 at 3:26 AM #8337Nice work Simon
1 user thanked author for this post.
05/30/2016 at 11:00 AM #8354Hi Simon, I looked at the code but I can’t understand the strategy. Could you please explain a bit how trading decisions are made accordingly to it? Because actually variables are only set by your optimization range values, they are not set by anything else 🙂
in the last code here: http://www.prorealcode.com/topic/state-space/#post-8318
05/30/2016 at 5:43 PM #8399Hi Nicolas! Thanks!
The system is described here:
http://link.springer.com/chapter/10.1007/978-88-470-2706-0_33
And look inside.I don’t know so much about the system, or understand it so well 🙂 I just know that it performs good in backtest, and now when I test it on demo-account (only about one month, I have tested it on OMX30 (almost like DAX)).
I’m a bit impressed with te system.
It need to be optimized now and then, my suggestion would be once a year, or every third month, or every single month. I don’t know so much about the system yet.
But I actually likes the fact that it needs to be optimized from now and then, cause I think that the market changes over time and and that a system needs to be adaptable. I don’t see it as curvefitting, especially when many results in optimizing gives good results. But I’m not sure 🙂I would like to be able to run it on smaller timeframes just to be able to optimize SL (because lack of look inside bar), and see what works best.
All the best! -
AuthorPosts