State Space

Viewing 3 posts - 16 through 18 (of 18 total)

Author

Posts
05/30/2016 at 8:43 PM #8432 quote
Nicolas
Keymaster

Master
State space reminds me of brownian motion.

They are no magic behind the results the optimizer give you, it’s just optimization of your 7 variables I believe. But it makes me curious so I did an indicator with the best optimized values of these 7 variables. As far as I can see the return value is so high that PRT cannot give its real because it is out of scale.. This is very weird..

This is the indicator code:
```
d1=0
d2=0
d3=1
d4=-1

n1=1
n2=1
n3=1
n4=0

dx1=-d1*x1-d2*x2-d3*x3-d4*x4+close
dx2=x1
dx3=x2
dx4=x3

x1=x1+dx1
x2=x2+dx2
x3=x3+dx3
x4=x4+dx4

StateSpace=n1*x1+n2*x2+n3*x3+n4*x4

RETURN statespace
```
.. but as you can see in the attached pictures, the “statespace” variables give almost dumb values which tends to signify that optimize it is like random coin toss.

Sorry, but I don’t think that there is something very interesting here.

statespace-1.png statespace-2.png
05/31/2016 at 6:19 AM #8454 quote

simon_json
Participant

Average

Thanks Nicolas! Please test it on daily TF. Do you think it possibile to get the exact same result on 5min TF by changing some code (on the same period of time, one year for example).

05/31/2016 at 8:02 AM #8460 quote

Nicolas
Keymaster

Master

As I told you, there is nothing here. So you can optimize on every timeframe you’d like, you can have good results, like almost any kind of strategy. Sorry.
Author

Posts

Viewing 3 posts - 16 through 18 (of 18 total)

You must be logged in to reply to this topic.

Login your account

State Space

ProOrder: Automated Strategies & Backtesting

New Reply

Author

simon_json @simon_json Participant

Summary

This topic contains 17 replies,
has 5 voices, and was last updated by Nicolas
9 years, 8 months ago.

Topic Details

Forum:	ProOrder: Automated Strategies & Backtesting
Language:	English
Started:	05/20/2016
Status:	Active
Attachments:	No files

Loading...