OmniTrader Forum OmniTrader Forum
forums calendars search
today this week
 
register logon control panel Forum Rules
You are currently browsing as a guest.
You should logon to access more features
A Self-Moderated Community - ALL MEMBERS, PLEASE READ!
Vote for Members who contribute the most to your trading, and help us moderate content within the Forums.


  Current location        Thread information  
OmniTrader Archives
OmniTrader 2014 Technical Support
Confirm Block Problem
Last Activity 10/7/2017 1:07 AM
14 replies, 1498 viewings

Jump to page : 1
Now viewing page 1 [25 messages per page]
 
back reply
Printer friendly version

^ Top
Vinay

Legend
100100100100252525
Posts: 479

Joined: 12/9/2011
Location: Planet Earth

User Profile
 
Subject : Confirm Block Problem
Posted : 9/18/2014 11:45 AM
Post #33553


Can any one enlighten me why the Confirm Block is not allowing the Signal to pass through it under the following conditions?





In the above case although the Signal was fired on 28/07/2014 and Confirming Signal was fired on 31/07/2014 which is within the tolerance period of 5 days, no Signal was passed through Confirm Block.





In this case both Signal and Confirming Signals were fired on 01/08/2014 but still the Signal was not passed through the Confirm Block.


[Edited by Vinay on 9/18/2014 11:56 AM]

Attached file : 1.jpg (49KB - 556 downloads)
Attached file : 2.jpg (45KB - 548 downloads)

^ Top
Vinay

Legend
100100100100252525
Posts: 479

Joined: 12/9/2011
Location: Planet Earth

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 9:36 AM
Post #33554 - In reply to #33553

On further investigation I have found that when more than one Strategy in the Confirm Block fires confirming Signal within the tolerance limit, OT shows as having fired the Signal when the last of the Signal is fired. In case of the RTM Strategies it means that it enters the trade at the best possible rate (at least on the Voteline) which is not possible except on hindsight. So effectively in the present case OT is cheating and presenting a very rosy picture.

In the case detailed in the previous post, I finally got the Signal on 04/08/2014 which is the best entry point in the present case. Please see the picture below to understand what I mean.






[Edited by Vinay on 9/19/2014 9:37 AM]

Attached file : Chart.jpg (147KB - 538 downloads)

^ Top
Vinay

Legend
100100100100252525
Posts: 479

Joined: 12/9/2011
Location: Planet Earth

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 9:38 AM
Post #33555 - In reply to #33553

Here is the screenshot of the Confirm Block.





[Edited by Vinay on 9/19/2014 9:40 AM]

Attached file : Confirm.jpg (182KB - 537 downloads)

^ Top
Vinay

Legend
100100100100252525
Posts: 479

Joined: 12/9/2011
Location: Planet Earth

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 9:55 AM
Post #33556 - In reply to #33553

Interestingly when I removed the T3S3 system from the Confirm Block which fired the confirming signal the last and which went to Voteline, the Signal on Voteline shifts to left by one bar when another system T3S1 fired the confirming Signal.

Moreover when I removed T3S1 also then the Signal on Voteline shifts to left by one bar once again when the confirming Signal was fired by VBX-3.

It is clear that the Signal on the Voteline changes with each passing day if there are more than 1 system in the Confirm Block firing confirming signals during tolerance period.

I wasted hundreds of hours first to develop this seemingly "amazing" strategy and then to figure out what's wrong with it. It is surprising that even after so many years of its existence the Confirm Block have such serious bug which went undetected or may be detected but not fixed.


P.S. Barry please move this post to Technical Support Forum.


[Edited by Vinay on 9/19/2014 10:05 AM]

^ Top
Jim Dean

Idol
200050010010010025
Posts: 2831

Joined: 9/21/2006
Location: L'ville, GA

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 10:08 AM
Post #33557 - In reply to #33555

I agree that this appears to be a bug ... your doc's really illustrate it well.

There has always been ambiguity/confusion regarding Tolerance inputs since the labels are too terse. The help info doesn't help (me) much since it offers no explanation of how "before" and "after" differ:
Confirmation Tolerance—Specifies the number of bars that OmniTrader will look back in order to accept a signal to be used for confirmation purposes. This refers to the signals generated by the Confirmation block as well as any signals passed to the block from a Secondary Signal Line (see Primary and Secondary Signal Lines).


Here is my understanding of what the two Tolerance inputs mean ... Barry, if this is not right, please correct me on this:

1. "Before Signal" means if the CONFIRM system(s) signal occurs BEFORE the actual VOTED system(s) signal, then still "counts" as confirmation as long as the voted-systems signal is no more than this many bars after the confirm-systems signal. IN THIS CASE, the trade would enter when the VOTED systems signal hits.

2. "After Signal" means if the CONFIRM system(s) signal occurs AFTER the actual VOTED system(s) signal, then the voted-systems signal still "counts" as long as the confirm-systems signal is no more than this many bars after the voted-systems signal. IN THIS CASE, the trade would enter when the CONFIRM systems signal hits.

The bug here is that #2 should apply (confirm sig is within 5 bars after the vote sig) ... but the ENTRY should be on the CONFIRM-sig bar, not on the voted-sig bar. As Vinay suggested ... the way it's working now is "crystal-ball cheating".

At least, that's how I understand it.


[Edited by Jim Dean on 9/19/2014 10:10 AM]

^ Top
Vinay

Legend
100100100100252525
Posts: 479

Joined: 12/9/2011
Location: Planet Earth

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 10:33 AM
Post #33558 - In reply to #33553

Jim... I got suspicious when this simple strategy produced 7700% ROI with reasonable DD without any margin over a period of 9 years. It shows PPT of 2% with an HR of 80%. I for once thought that I have found the proverbial "Holy-Grail" strategy. But as they say if it's too good to be true, it rarely turns out to be true.

With each new bug discovery I am losing confidence in the reliability and accuracy of OT.

^ Top
Jim Dean

Idol
200050010010010025
Posts: 2831

Joined: 9/21/2006
Location: L'ville, GA

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 10:44 AM
Post #33559 - In reply to #33558

My response to finding a bug (which occurs very rarely, in light of the plethora of features) is to stop using that feature until it is fixed.

In this case, just stop using the "After Signal" input.

All complex programs have bugs. The key difference is ... how easy is it to find them, to report them, and get a fix done for them? Nirvana is usually very responsive, for bugs that affect many people and situations.

[Edited by Jim Dean on 9/19/2014 10:46 AM]

^ Top
Barry Cohen

Sage
5000100100100252525
Posts: 5395

Joined: 1/19/2004

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 4:12 PM
Post #33560 - In reply to #33558

Vinay, please post or email me the strategy & any OmniLanguage files needed to run it & I'll be happy to take a look.
^ Top
Jim Dean

Idol
200050010010010025
Posts: 2831

Joined: 9/21/2006
Location: L'ville, GA

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/19/2014 4:22 PM
Post #33561 - In reply to #33560

Barry:

Are my elaborations/definitions about what Before and After mean, and how they are supposed to work correct, in the earlier post?

[Edited by Jim Dean on 9/19/2014 4:23 PM]

^ Top
Vinay

Legend
100100100100252525
Posts: 479

Joined: 12/9/2011
Location: Planet Earth

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/22/2014 7:54 AM
Post #33563 - In reply to #33553

Barry... I have mailed you the Strategy, VBA System as well as the MetaStock Data files for the Symbol used in the screenshots posted by me.

Please do the needful in the matter and get it fixed as soon as possible.


[Edited by Vinay on 9/22/2014 7:55 AM]

^ Top
Barry Cohen

Sage
5000100100100252525
Posts: 5395

Joined: 1/19/2004

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 9/22/2014 6:06 PM
Post #33568 - In reply to #33554

Thanks for sending everything that I needed to see the issue, Vinay - it helps alot. I checked it out. There are 2 signals produced from your system. On 7-28 & on 8-1 & that's causing this strangeness. If you change your "After Signal" tolerance in the Confirm block to 2 instead of 5, you should see the VBX-3 signal make it to the voteline. I'll report the issue, but I can't say when it will be addressed.

Are my elaborations/definitions about what Before and After mean, and how they are supposed to work correct, in the earlier post?


Yes Jim.
^ Top
Vinay

Legend
100100100100252525
Posts: 479

Joined: 12/9/2011
Location: Planet Earth

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 7/20/2015 9:19 AM
Post #35620 - In reply to #33553

It has been a long time since this issue was reported. Till this issue is fixed, we can not use this important feature in both OT & VT because it is giving misleading results.

Kindly get it fixed on high priority.


^ Top
John J

Veteran
10010025
Posts: 233

Joined: 8/3/2010
Location: Leduc, AB

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 2/4/2016 10:57 PM
Post #36007 - In reply to #33553

Hi Barry,

Has this issue been corrected yet?
^ Top
Barry Cohen

Sage
5000100100100252525
Posts: 5395

Joined: 1/19/2004

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 2/10/2016 10:01 AM
Post #36038 - In reply to #36007

I'm sorry, but it has not been addressed yet.
^ Top
Jim Dean

Idol
200050010010010025
Posts: 2831

Joined: 9/21/2006
Location: L'ville, GA

User Profile
 
Subject : RE: Confirm Block Problem
Posted : 2/10/2016 10:09 AM
Post #36041 - In reply to #36038

Hi, Barry

If it's going to be a while, could you ask the folks simply to gray out the "After" box in the meantime, to keep people out of trouble?

Unless they read this thread, or watch the ROI's like Vinay did, they could be misled by the results into real-$$ losses.

[Edited by Jim Dean on 2/10/2016 10:10 AM]

Jump to page : 1
Now viewing page 1 [25 messages per page]
back reply

Legend    Action      Notification  
Administrator
Forum Moderator
Registered User
Unregistered User
E-Mail this thread to a friend
Toggle e-mail notification


Nirvana Systems
For any problems or issues please contact our Webmaster at webmaster@nirvsys.com.