| Pages:
1
..
55
56
57
58
59
..
98 |
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
Quote: Originally posted by admin  | | Quote: Originally posted by Daniel UK1 | Hi Nick,
I believe this wording at IB creates the issue i am mentioning.
So as far as my understanding, you can close out positions, but you can not open a new trade, "within 5 days to expiry of closing contract"
And i Believe IQ rolls 3 days before contract ends for CL.
"Additionally, beginning five business days prior to expiration, only margin-reducing transactions in expiring month contracts will be accepted for
oil-based products eligible to trade at negative prices."
https://www.interactivebrokers.com/en/index.php?f=deliveryEx... |
I think this is correct. in ts the correct symbol for cl (via ib) is |
Hi Peter, Yes one can solve this by using custom contract you show in TC, we can also do this in MC, but this is more about if one wants to use the
adjusted or non adjusted symbol IQ qcl c or just qcl,
I am pointing towards the regulation that no new positions is allowed in CL if trading with IB, after -5 days before contract ends. And IQ rolls 3
days for CL as far as i know, meaning that iif you wait for IQ rolls, you have 2 days where your orders for new positions will be rejected by IB if
you waited to roll when IQ rolls.
This is my understanding based on the text on IB page, in my link above.
Perhaps someone else have a different or better understanding than me about this.
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
Hello, the information is difficult to understand, sorry.
Can you give the name of the macro that makes the 300 and then the families?
Thanks.
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
| Quote: Originally posted by REMO755 | Hello, the information is difficult to understand, sorry.
Can you give the name of the macro that makes the 300 and then the families?
Thanks. |
Remo, if i understand you correct, you want to know the macro that places the system top 300 OOS in B and then top families in in C, you will find
that in macro calledNQ_is2016 etc...
You then if you follow peter, change top 300 in B to no TRD Dates,then look at the OOS results, and based on that WF the systems you find
interesting.. or you do this only with the ones in C.. As you like..
Wf is manual done of the ones you want to WF, remember to change dates back and Nth day to all before WF thought.
I think peter will fill in if something is missing.
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
Thanks Daniel,
I have to understand this part, you say:
Remember to change the dates back and N-day for everyone before WF thought.
I do not understand this, what is it for?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by REMO755 | Thanks Daniel,
I have to understand this part, you say:
Remember to change the dates back and N-day for everyone before WF thought.
I do not understand this, what is it for? |
the macro might leave dates say from 20200630 to 20210631
This is not what you want to wf.
make sure all dates are used in the wf, and nth is on all. (likely to be the case)
when you look at a system on the graph, it should go from 1997/2000/2007 etc to
say 2020 before wf is done
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by REMO755 | Thanks Daniel,
I have to understand this part, you say:
Remember to change the dates back and N-day for everyone before WF thought.
I do not understand this, what is it for? |
the macro might leave dates say from 20200630 to 20210631
This is not what you want to wf.
make sure all dates are used in the wf, and nth is on all. (likely to be the case)
when you look at a system on the graph, it should go from 1997/2000/2007 etc to
say 2020 before wf is done |
Sorry, I don't understand this part well
What is the difference between All, NoTrd or Trd?
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by REMO755 | Thanks Daniel,
I have to understand this part, you say:
Remember to change the dates back and N-day for everyone before WF thought.
I do not understand this, what is it for? |
the macro might leave dates say from 20200630 to 20210631
This is not what you want to wf.
make sure all dates are used in the wf, and nth is on all. (likely to be the case)
when you look at a system on the graph, it should go from 1997/2000/2007 etc to
say 2020 before wf is done |
Sorry, I don't understand this part well
What is the difference between All, NoTrd or Trd?
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
They are used to configure the training periods inside and outside the sample.
Trd: it is a configuration that tells you that between the dates you build the system you are going to trade, that is, you train the system with that
data, they are sample periods.
Ntrd: it is a configuration that tells you that between the dates that you build the system you will not trade, that is, they will be out of sample.
ALL: are the periods within the sample and outside the sample.
Thanks received (2):
+1 loclhero at 2021-07-06 12:10:07 +1 REMO755 at 2021-07-02 14:32:44
|
|
|
bud67
Junior Member
Posts: 12
Registered: 22-4-2019
Member Is Offline
Mood: No Mood
|
|
Sometimes when i do a WF @BestWorker I get exceptions like these:
It's only happening with the ES example not the NQ.
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
Hi Peter
I give you the results of the complete process of building a NQ system.
Do you detect an error in the process? Can you audit it for me?
In terms of systems, I think it is good for us, very unstable. It is right. Do you have any guide to interpret wfo results? I know you have a video,
but don't you have a guide?
     
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
Peter all those zeros that are parameters, why are they zero?
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
Hello,
Is it possible to do this in GSB and then replicate in TS?
Hours 03:30 - 7:30
EC data 15. Regular. Exchange
MOC @ 730
GSB everything is perfect, departures 7.30, ok, when I import code in TS the departures are not at 7.30, the departures are at the close of the day.
That's wrong? It's a mistake? Something badly established? Where is the problem?
|
|
|
Carl
Member
Posts: 342
Registered: 10-5-2017
Member Is Offline
Mood: No Mood
|
|
I think your TS session differs from the session used in GSB.
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
Hi carl
It is the usual time of the session.
The problem is that MOC 07.30 does not coincide with the closing time of the day, so TS does not replicate the code well.
I think it is a bug, or maybe I need to configure it differently.
I await Peter's response.
I am checking that when the session time is greater than the closing time of the MOC, TS does not close in MOC and GSB does.
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
| Quote: Originally posted by REMO755 |
Hi carl
It is the usual time of the session.
The problem is that MOC 07.30 does not coincide with the closing time of the day, so TS does not replicate the code well.
I think it is a bug, or maybe I need to configure it differently.
I await Peter's response.
I am checking that when the session time is greater than the closing time of the MOC, TS does not close in MOC and GSB does.
|
Hi carl
Another example Carl, can you do the test?
Data: regular schedule ES of 30 minutes
Times: 08.30-9.00
MOC (market at close): TRUE
Session Moc @ 1230
You will see how GSB closes at 12.3 and TS closes at 16.00
Another example Carl, can you do the test?
Data: regular schedule ES of 30 minutes
Times: 08.30-9.00
MOC (market at close): TRUE
Session Moc @ 08.00-1230
You will see how GSB closes at 12.3 and TS closes at 16.00
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
| Quote: Originally posted by portfolioquanttrader2020 | Hi Peter
I give you the results of the complete process of building a NQ system.
Do you detect an error in the process? Can you audit it for me?
In terms of systems, I think it is good for us, very unstable. It is right. Do you have any guide to interpret wfo results? I know you have a video,
but don't you have a guide? |
Hi 2020, I am not sure its possible to see if your process is off, from what you have sent. As long as you have used Peters dates for getting the
main indicators, and then his macros for stats, and based your decisions along the way from a large quantity of systems, and then picked whole
families based on OOS (not single systems )and fitness relative to all systems in B, AND then picking systems based on that together with WF
stability, you should be pretty ok..... you wf example would not be on top of my list... me personally i would have liked better stablity of a
system, also important would be how rest of your family looks like, are they stable etc  .. to feel better with a system you can test on other markets, other timeframes, etc, try to break it
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
Peter, on what do you base to choose the size of the stop on the Nasdaq?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
$2000 is likely my stop for most markets.
WHen a system is built you can manually optimize the stop.
last time I did this was $1900 the best, but I figure I like to go a little larger
as out of sample I feel I will get little better results.
Also many markets are extremely volatile right now, and tight stop will hurt profits
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
Hi peter
I understand that in order to create the methodology for each asset, you do different tests, and you see what works best, and the most robust. And
once, you find the most robust and what works best, you write the methodology of each asset.
It is right?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by portfolioquanttrader2020 | Hi peter
I understand that in order to create the methodology for each asset, you do different tests, and you see what works best, and the most robust. And
once, you find the most robust and what works best, you write the methodology of each asset.
It is right? |
Correct. Time consuming to do each market, but then very easy for those that follow using the same settings
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
Peter, do you think it is possible to know what is the best time of entry and exit in an asset, to create a robust system? Do you think there are
robust hourly advantages? In GSB can you investigate the hourly advantages?
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
2020, one idea is to arrive to your initial idea forbest session time by exporting data and look at volume at time.. that would give you an idea,
peter has created an indicator for that i believe and kindly shared it,.... after that yuo can use automation to toggle trough your time sessoins,
and based on that you can draw you pick what you think is best, and do further testing.... or perhaps you question is not about session times but
more entry and exit time?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
@portfolioquanttrader2020 I think the latest copy of gsb automation allows you to shift the entry time, however in nearly all cases the best entry
time is the beginning of the session time used.
THe more important issue is whats the session time.
for es ym nq rty etc that's really simple. GC,DAX much more complex as there can be multiple valid session times
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
Thanks Daniel
My question refers to hour filters, to look for simple advantages, hourly, I think they are seasonal calls.
Create systems only with hourly filters and an indicator for example.
|
|
|
REMO755
Member
Posts: 181
Registered: 11-4-2021
Member Is Offline
|
|
Hello
Default secondary filter ------- Auto mode vs genetic mode, what is the difference?
|
|
|
| Pages:
1
..
55
56
57
58
59
..
98 |
![[*]](images/davis/default_icon.gif){kind=icon}
