| Pages:
1
..
3
4
5
6
7
..
17 |
uhrbi
Junior Member
Posts: 24
Registered: 21-12-2017
Member Is Offline
Mood: No Mood
|
|
What do you think of a feature to only enter a new trade when there is no position? (Marketposition = 0)
Regards,
uhrbi
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by uhrbi  | What do you think of a feature to only enter a new trade when there is no position? (Marketposition = 0)
Regards,
uhrbi |
GSB currently can reverse a position which in my opinion is a great idea.
It stops bigger losses. It doesnt not pyramid.
You could modify your GSB code to remove this as a test to see for yourself.
|
|
|
uhrbi
Junior Member
Posts: 24
Registered: 21-12-2017
Member Is Offline
Mood: No Mood
|
|
I see your point.
I was coming from the idea to add "truisms" later on in TS, and when I do that, for let's say the long trades, it can affect the short trades, because
there might not be a position to reverse in the first place.
|
|
|
curt999
Junior Member
Posts: 51
Registered: 24-7-2017
Member Is Offline
Mood: No Mood
|
|
try building strategies that are only long short then combining them makes it easier to see which truisms affect what..but you can really do this in
the gsb strategy code too just add..create truisms independtly for long or short then apply then in the line below
If decision = 1 And sfDecision = 1 and "your custom condition goes here" Then enter long or short
\
|
|
|
uhrbi
Junior Member
Posts: 24
Registered: 21-12-2017
Member Is Offline
Mood: No Mood
|
|
Thank you for the advice, I'll try that.
Have you achieved good results with building strategies separately for long and short and combining them afterwards?
|
|
|
curt999
Junior Member
Posts: 51
Registered: 24-7-2017
Member Is Offline
Mood: No Mood
|
|
yes i have.plus when you design this way you end up with a non symmetrical strategy with different conditions for long and short..its more work to do
it this way however compute time etc..another reason i like to do this is if the overall strategy is peforming poorly but its only one direction not
both you arent back to square one..
|
|
|
dpetaz
Junior Member
Posts: 2
Registered: 24-1-2018
Member Is Offline
Mood: No Mood
|
|
GPU
Hello Peter, since GSB needs a lot of CPU power, did you think to increase the computational speed using also the GPU? According to some documents I
read, this could improve performance 10x to 60x, for high end video card.
|
|
|
rws
Member
Posts: 114
Registered: 12-6-2017
Member Is Offline
Mood: No Mood
|
|
Here is another aplication that was able to improve speed 100-1000X doing this and his open code in Lazarus was already very fast.
This was only price patterns but it could work for any type of scripts especially because GA is a parallel calculation.
http://mechanicalforex.com/kantu-system-generator
| Quote: Originally posted by dpetaz | | Hello Peter, since GSB needs a lot of CPU power, did you think to increase the computational speed using also the GPU? According to some documents I
read, this could improve performance 10x to 60x, for high end video card. |
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by rws | Here is another aplication that was able to improve speed 100-1000X doing this and his open code in Lazarus was already very fast.
This was only price patterns but it could work for any type of scripts especially because GA is a parallel calculation.
http://mechanicalforex.com/kantu-system-generator
| Quote: Originally posted by dpetaz | | Hello Peter, since GSB needs a lot of CPU power, did you think to increase the computational speed using also the GPU? According to some documents I
read, this could improve performance 10x to 60x, for high end video card. |
|
This wont apply to GSB, as all the maths is cached. Hence GSB does very little maths. The next beta build of GSB (43.13) is going to have option for
just over 20% increase in speed. It will be in the private forum only.
GSB cloud also gives unlimited scope for speed improvement too.
|
|
|
rws
Member
Posts: 114
Registered: 12-6-2017
Member Is Offline
Mood: No Mood
|
|
I understand you have other priorities and rewriting software for supporting a GPU is a lot of work.
Comparing OHLC data is even less math and it was 100-1000X faster.
Especially in parallel processing which could be a great portion of
Wf in GSB and GA a GPU shines.
http://www.nvidia.com/object/what-is-gpu-computing.html
Nvidia stock is up so much because their GPU chips are so efficient in machinelearning and parallel processing.
Even VMware uses GPUs now and that is running a complete operating system.
Google tensorflow runs very well on GPU.
https://www.tensorflow.org/programmers_guide/using_gpu
|
|
|
curt999
Junior Member
Posts: 51
Registered: 24-7-2017
Member Is Offline
Mood: No Mood
|
|
heh no nvdia stock is up so much because people are buying up the gpu to mine with has nothing to do with machine learning
|
|
|
rws
Member
Posts: 114
Registered: 12-6-2017
Member Is Offline
Mood: No Mood
|
|
Sure also because of bitcoin mining.
https://www.forbes.com/sites/aarontilley/2017/05/09/nvidia-1...
http://fortune.com/2017/12/05/ibm-ai-chip-nvidia/
http://fortune.com/2016/11/10/nvidia-earnings-stock-machine-...
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
We will look into this, but not in a hurry. Current GSB can be optimized 1.5 to 3 times faster we think. Only after this is done can we look into GPU.
There is a post in private forum to get bit over 20% faster too.
But priorities are WF on workers, More secondary filters and other stop exits
|
|
|
Carl
Member
Posts: 342
Registered: 10-5-2017
Member Is Offline
Mood: No Mood
|
|
GSB already has the possibililty to switch off/on the "secondary filter".
I would like to have the possibility to switch off the "primary filter", to see what kind of secondary filter works best on a particular ticker and
session.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Carl | GSB already has the possibililty to switch off/on the "secondary filter".
I would like to have the possibility to switch off the "primary filter", to see what kind of secondary filter works best on a particular ticker and
session. |
I dont think thats going to work, but im not sure.
You can turn off possible all but one primary filters.
I still think you going to need both pri and secondary for testing.
Much work to be done on Secondary filters in the next few months
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Carl | GSB already has the possibililty to switch off/on the "secondary filter".
I would like to have the possibility to switch off the "primary filter", to see what kind of secondary filter works best on a particular ticker and
session. |
Bottom line is in most markets nothing comes close to closeD>close filter.
|
|
|
Carl
Member
Posts: 342
Registered: 10-5-2017
Member Is Offline
Mood: No Mood
|
|
Hi Peter,
Thanks for version 44.09 and adding the WF data into the TS script.
Would be nice if the IS, OOS and VAL date ranges were added to the TS script.
Thanks.
|
|
|
cyrus68
Member
Posts: 171
Registered: 5-6-2017
Member Is Offline
Mood: No Mood
|
|
The Sharpe ratio is already calculated in GSB and is reported in the performance stats. It would be extremely useful to have it available in the
panel, alongside the other performance metrics, so that we could rank systems according to their Sharpe-ratio score.
The Sharpe ratio calculates return, per unit of risk - where risk is defined as the standard deviation of returns. It is the best overall measure of
the quality of a system.
It would also be useful to have the Sharpe ratio reported in the WF results, for the OOS and Current curves.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by cyrus68 | The Sharpe ratio is already calculated in GSB and is reported in the performance stats. It would be extremely useful to have it available in the
panel, alongside the other performance metrics, so that we could rank systems according to their Sharpe-ratio score.
The Sharpe ratio calculates return, per unit of risk - where risk is defined as the standard deviation of returns. It is the best overall measure of
the quality of a system.
It would also be useful to have the Sharpe ratio reported in the WF results, for the OOS and Current curves.
|
This seems a reasonable idea. I'm open to doing this, but don't want to clutter the GUI more.
You can use fitness of sharp ratio but I prefer NP*AT fitness.
What about making one field user definable? Ie default my be % profitable, but it can be changed to any of the other metrics.
|
|
|
rws
Member
Posts: 114
Registered: 12-6-2017
Member Is Offline
Mood: No Mood
|
|
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2745220
| Quote: Originally posted by cyrus68 | The Sharpe ratio is already calculated in GSB and is reported in the performance stats. It would be extremely useful to have it available in the
panel, alongside the other performance metrics, so that we could rank systems according to their Sharpe-ratio score.
The Sharpe ratio calculates return, per unit of risk - where risk is defined as the standard deviation of returns. It is the best overall measure of
the quality of a system.
It would also be useful to have the Sharpe ratio reported in the WF results, for the OOS and Current curves.
|
|
|
|
rws
Member
Posts: 114
Registered: 12-6-2017
Member Is Offline
Mood: No Mood
|
|
I have no opinion on this but just red it a while ago.
| Quote: Originally posted by rws |
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2745220
| Quote: Originally posted by cyrus68 | The Sharpe ratio is already calculated in GSB and is reported in the performance stats. It would be extremely useful to have it available in the
panel, alongside the other performance metrics, so that we could rank systems according to their Sharpe-ratio score.
The Sharpe ratio calculates return, per unit of risk - where risk is defined as the standard deviation of returns. It is the best overall measure of
the quality of a system.
It would also be useful to have the Sharpe ratio reported in the WF results, for the OOS and Current curves.
|
|
|
|
|
zdenekt
Junior Member
Posts: 37
Registered: 1-10-2017
Location: Czech Republic
Member Is Offline
Mood: No Mood
|
|
It would be nice to have a "Win/Loss ratio" as fitness.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
It can be done.
My belief is np*at is best fitness, but performance metrics is what you want to be very tight.
Also the more metrics we calculate, the slower GSB gets.
ie if commission on fitness <> commission on results, gsb has to do all its maths twice. Slows GSB down about 12%
|
|
|
cyrus68
Member
Posts: 171
Registered: 5-6-2017
Member Is Offline
Mood: No Mood
|
|
Peter: I like the idea of a user definable field. To be able to switch between % profitable and the Sharpe ratio would be ideal.
Also, if it would be possible to have the Sharpe ratio displayed, after WF, for the OOS and Current curves. At the bottom of the parameters table
perhaps.
rws: I can't comment on the ssrn paper until I've read it.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by cyrus68 | Peter: I like the idea of a user definable field. To be able to switch between % profitable and the Sharpe ratio would be ideal.
Also, if it would be possible to have the Sharpe ratio displayed, after WF, for the OOS and Current curves. At the bottom of the parameters table
perhaps.
rws: I can't comment on the ssrn paper until I've read it. |
Assuming the programmer can do it ok, we can do this. There is a long list of small tweaks like this, so I will add it. Every few builds you probably
notice some little gui tweaks etc
|
|
|
| Pages:
1
..
3
4
5
6
7
..
17 |
![[*]](images/davis/default_icon.gif){kind=icon}
