Jump to content


Photo

New rule for AutoPilot ?


  • Please log in to reply
1 reply to this topic

#1 Clive Rich

Clive Rich

    Newbie

  • Members
  • Pip
  • 3 posts
  • LocationUK

Posted 21 October 2011 - 10:23 AM

All

At the European PKS User Meeting in Paris, users have suggested the inclusion of a business rule based on lamba z fit parameters, in the same way that the AUC Extrapolation rule currently exists.

 

Should this be put forward as a formal requirment for AutoPilot V3 ?

 

Clive



#2 Simon Davis

Simon Davis

    Advanced Member

  • Administrators
  • 1,316 posts

Posted 21 October 2011 - 06:15 PM

Hi Clive thanks for posting this comment, I think we will try to collate the user comments/requests from this session and circulate as a whole for review but in the mean time it's probably a good place to capture comments and suggestions.

 

  There is an existing related issue; QC_AP_456, which we could perhaps refine.

 

I think the general impression this morning was that users were interested in flagging/masking Lz dependent parameters when adjusted R-squared was less than some admin/user specified value e.g. 0.85. Much like we currently do with AUCinf%extrap.

 

I think the QC_AP_456 I refer to above is looking to qualify this by an additional check I have come across with various analysis protocols. in the past.

 

Flag lambda_z dependent estimates if thalf is not estimated over at least X half-lives where X is typically between 2 and 3 and would be set by the admin/user at run time.

 

 Simon.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users