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:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
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
RETURNstatespace
.. 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.
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).
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.
