| Pages:
1
..
19
20
21
22
23
..
47 |
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by Daniel UK1  | | Quote: Originally posted by admin | Just a reminder,
if you want to contribute to the CL research being done by letting me your your workers, then send me the your share keys
I will in turn give you the output
Here is an example of tests im doing
Im up to over 200 workers (thanks for those who have given me workers)
What im doing here is 4 tests each on 7 different secondary filters.
Im also doing similar tests, some with 40 inidicators, some with 100 indicators
some two pass using green / green and some with green.orange/green etc
|
I assume that you are figuring out for your process if its best to use a very low amount of indicators, medium or all in order to do optimal SF.. this
is what your research in this specific test is trying to establish right ? |
yes, im trying numerous sf, 40 vs 100 indicators, green+orange,vs green,green
and some various bar intervals
Thanks received (1):
+1 Daniel UK1 at 2020-07-27 09:04:48
|
|
|
DocBober55
Junior Member
Posts: 17
Registered: 18-6-2020
Member Is Offline
|
|
...
Thats covered in the last CL video
From GSB CL system thread:
...This system is the system shown at the end of the CL video." - [N.B. the link provided in the post
(https://www.youtube.com/watch?v=cNFvhNlzugY) leads me to an "unavailable" video]
I want to make sure I am looking at the most recent CL video, or at least the one you are referencing. What is the date? Is it the
CrudeOil_video4.14_unfinished that I have saved, or
another one?
Thanks
|
|
|
DocBober55
Junior Member
Posts: 17
Registered: 18-6-2020
Member Is Offline
|
|
Peter,
If I give you my share key but then want to try to use GSB myself to develop a system, will it be possible for me to do so if you are simultaneously
conducting your research?
I only have 16 GB of RAM and I believe one worker.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by DocBober55 | Peter,
If I give you my share key but then want to try to use GSB myself to develop a system, will it be possible for me to do so if you are simultaneously
conducting your research?
I only have 16 GB of RAM and I believe one worker. |
I appreciate the offer, but a low end machine is not worth the effort for us both,
esp as we cant both use the worker at the same time.
I might be finished CL runs today, depends what I find in the results.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
next automation run, I want to test top 4,6,8,10 indicators, regardless of how many are green.
GSB needs a tweak to do this, so likely will be a day or two away.
Wild guess its 4 days processing to do this on 12 secondary filters.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Just giving an update on progress
I hired another programmer to convert stats from gsb export into excel
Im going to have to do all my tests due to format changes needed for macro, and human error in the config file
Will give an example here on what it looks like when tests are redone. Its going to be a day or so before they are ready
This is the setting that need to be used with the excel spread sheet.
The excel macro wont be free, but it will be affordable
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Here is a output of the summary field in the excel macro
Shown is the results of all SF for the specific test setup.
Im going to redo all results
but Im going to do benchmark on 9 SF using 15.30 min bars
Then compare to
some other bar intervals
Force min of 4,6,8,10 indicators
try #indicators of 2,4,5 (all tests done on 3 so far)
With the 200 or so workers I have from my own cloud, and the generous GSB users
I can do about 2 tests per day.
ie 9 sf on say force4 indicators & 9 indicators on say force6 indicators for example.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
update on CL
ive done the same test with 3,4,5 indicators#
tests with 4,6,8,10 indicators forced
need to redo min of 4 indicators, and the default of 15.30 with min 4 indicators greenorange,green
This will take about 2 more days.
Results all nice and clean in that everything is 100% automated. Even the excel work
Thanks again to those who contributed workers
Thanks received (1):
+1 OUrocketman at 2020-08-18 22:50:08
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
This is CL tertiary filters first pass
There is no TS code done yet
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
here is pass2
Once tf is coded, then we can add all the new price patterns etc
Thanks received (1):
+1 moresi522 at 2020-09-22 05:55:53
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Is anyone interested in settings that enter ES after the close of day, and exits at day session moc?
This is not recommend if you are new to GSB.
There is a clear market bias for long in this, but this is going to take a fair bit of work.
While im getting ok OOS results (quick tests) im not getting any improvement with 1 pass indicator test (unusual)
GSB also needs a new feature to maximise this.
|
|
|
Wingnuts
Junior Member
Posts: 5
Registered: 20-4-2020
Member Is Offline
|
|
Definitely interested.
Historically, the SPX has made ALL its gains in the US overnight session, and only broken even during the day session.
I already trade a non-GSB strategy to exploit this, but it uses a big stop loss and smaller target, so it's a bit lob-sided and uncomfortable to
trade.
The US overnight session should be rich pickings, and great for those of us outside the US who can watch the charts during those hours.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Wingnuts | Definitely interested.
Historically, the SPX has made ALL its gains in the US overnight session, and only broken even during the day session.
I already trade a non-GSB strategy to exploit this, but it uses a big stop loss and smaller target, so it's a bit lob-sided and uncomfortable to
trade.
The US overnight session should be rich pickings, and great for those of us outside the US who can watch the charts during those hours.
|
I will publish monday.
Thanks received (4):
+1 LucaRicatti at 2020-09-20 17:23:35 +1 Wingnuts at 2020-09-19 17:51:01 +1 bizgozcd at 2020-09-19 10:35:04 +1 Carl at 2020-09-19 07:23:02
|
|
|
portfolioquanttrader2020
Member
Posts: 269
Registered: 30-4-2020
Member Is Offline
|
|
Hello
After passing the WFO to an overnight system in GSB, when copying and pasting the script in TradeStation it does not appear in the parameters.
Besides, there are many traders and the BMO is very low.
How can I solve these problems?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
here is the es overnight macro that will load the settings.
Note as this is very early stages of building using this method, I have training and test at 50%
Attachment: Login to view the details
 
Attachment: Login to view the details
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by portfolioquanttrader2020 | Hello
After passing the WFO to an overnight system in GSB, when copying and pasting the script in TradeStation it does not appear in the parameters.
Besides, there are many traders and the BMO is very low.
How can I solve these problems? |
If this is a miss-match It is most likely - but not always human error.
check the start date of the contract, bar interval. (session time is 24/7 in this case)
you will not set parameters in ts, as we use vars instead of inputs.
This is deliberate. I got about 1 user per week with missmatch.
if you have system 1 in ts, and copy the code of system2 into your ts code, the inputs of system1 are remembered, not system2. This will cause a
missmatch.
the solution is delete the ts code in ts editor, hit f3, then paste the code of system2
You can wf the code in gsb. much faster from human and cpu time.
I would however move this reversal exit from ts.
In a future (soon i hope) build of gsb, this will be an option
I think you are better of with no reversal exit when you have no secondary filter
If stuck we can do a teamviewer session.
Short term es overnight systems will be best for users who know gsb and ts well.
Im not sure what bmo means
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
I made a simple discovery today.
I built using a data2, and got slightly better stats that data1.
When I looked at the top systems in top families of the top 250 systems, data2 was used only twice. I read into this that genetic algorithms see data2
as not worth using.
The logic behind 50,000 systems -> 250 -> familes is if its not common, its generally not good.
there were 26 systems here, 3 data streams,so 2/72 possible data streams used data2
If it was good, it would be a lot more.
done by right click the systems.....
Thanks received (2):
+1 bizgozcd at 2020-09-28 07:03:41 +1 OUrocketman at 2020-09-26 21:10:49
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
There is a critical TS 9.5 bug on @GC. Im using exchange time 700 to 1330.
Before march 2020, we get bars before 700am
On local time the issue is fixed.
However systems made in GSB on exchange time work well. Local time work like crap in GSB. (there should be zero difference)
GSB systems will not match ts if they are made on exchange time in TS.
If systems made in gsb on local time, match with ts is perfect.
I want to compare this to MC / iqfeed
Can someone send me gc.1minute bars
time 700 to 1330 exchange time, exported in exchange time and a file in local time
dates from nov 1 2006 till today
Im interested if ts 10 has the same issue, but prefer to try mc first
Later note.
ts 10 has identical issue
This is the date at which the bug shows.
|
|
|
DocBober55
Junior Member
Posts: 17
Registered: 18-6-2020
Member Is Offline
|
|
I do have MC and I read/plot TS data into MC. However, where I live, exchange time for GC is equivalent to local time for GC. Let me know if I can
be of any help.
|
|
|
DocBober55
Junior Member
Posts: 17
Registered: 18-6-2020
Member Is Offline
|
|
I do have MC and I read/plot TS data into MC. However, where I live, exchange time for GC is equivalent to local time for GC. Let me know if I can
be of any help.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by DocBober55 | | I do have MC and I read/plot TS data into MC. However, where I live, exchange time for GC is equivalent to local time for GC. Let me know if I can
be of any help. |
great, can you export 15 min data
700 to 1330 from 11.2006 till today?
|
|
|
Carl
Member
Posts: 342
Registered: 10-5-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by admin | There is a critical TS 9.5 bug on @GC. Im using exchange time 700 to 1330.
Before march 2020, we get bars before 700am
On local time the issue is fixed.
However systems made in GSB on exchange time work well. Local time work like crap in GSB. (there should be zero difference)
GSB systems will not match ts if they are made on exchange time in TS.
If systems made in gsb on local time, match with ts is perfect.
I want to compare this to MC / iqfeed
Can someone send me gc.1minute bars
time 700 to 1330 exchange time, exported in exchange time and a file in local time
dates from nov 1 2006 till today
Im interested if ts 10 has the same issue, but prefer to try mc first
Later note.
ts 10 has identical issue
This is the date at which the bug shows.
|
Hi Peter,
I have the same issue on other ticker symbols (TS 9.5 update 28).
Using session 0700-1500 on Exchange Time setting: but chart shows 0630-1500 between November and March. This issue returns every year in the chart.
The chart data between April and October is correct.
Update
On one of my laptops I have opened TS for the last time in July this year.
Opening TS offline now shows the correct price data between 2006 and July 2020 (TS 9.5 update 28).
Is it a bug in TS software or in the data?
Thanks received (2):
+1 admin at 2020-09-30 20:11:59 +1 Bruce at 2020-09-30 19:01:58
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Carl | | Quote: Originally posted by admin | There is a critical TS 9.5 bug on @GC. Im using exchange time 700 to 1330.
Before march 2020, we get bars before 700am
On local time the issue is fixed.
However systems made in GSB on exchange time work well. Local time work like crap in GSB. (there should be zero difference)
GSB systems will not match ts if they are made on exchange time in TS.
If systems made in gsb on local time, match with ts is perfect.
I want to compare this to MC / iqfeed
Can someone send me gc.1minute bars
time 700 to 1330 exchange time, exported in exchange time and a file in local time
dates from nov 1 2006 till today
Im interested if ts 10 has the same issue, but prefer to try mc first
Later note.
ts 10 has identical issue
This is the date at which the bug shows.
|
Hi Peter,
I have the same issue on other ticker symbols (TS 9.5 update 28).
Using session 0700-1500 on Exchange Time setting: but chart shows 0630-1500 between November and March. This issue returns every year in the chart.
The chart data between April and October is correct.
Update
On one of my laptops I have opened TS for the last time in July this year.
Opening TS offline now shows the correct price data between 2006 and July 2020 (TS 9.5 update 28).
Is it a bug in TS software or in the data?
|
THanks so much for this. Things like this could kill traders profitability, and we have no idea about it. Im going to document this is gsb docs.
Its a bug in ts session time logic. Solution is to use local time. What symbols have you seen it in? I sue local time normally, but developed cl on
exchange time. I need to check my data on this.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Later note. my cl data is ok. It it wasnt likely I would have to revisit all cl work again.
|
|
|
engtraderfx
Junior Member
Posts: 98
Registered: 15-10-2018
Member Is Offline
Mood: No Mood
|
|
Peter, just emailed some IQ/MC data, it only goes back to early 2007 though, don't know if others find this but i find IQ data a bit flakey in early
years around 2007/early 2008, gaps, low vol, spikes. I triy to get them to correct obvious spikes but generally exclude some of it from backtesting.
Don't remember having any issues with bars outside session time though. Regards, Dave
Thanks received (1):
+1 admin at 2020-09-30 23:12:49
|
|
|
| Pages:
1
..
19
20
21
22
23
..
47 |
![[*]](images/davis/default_icon.gif){kind=icon}
