Jump to content


Not estimable

  • Please log in to reply
2 replies to this topic

#1 Reader



  • Members
  • Pip
  • 2 posts

Posted 13 March 2017 - 08:45 PM



Could you please share your expirience with following issue? When I use such a model: Sequence+Period+Treatment+Group+Patient(Sequence) I get "Not estimable" in the results of ABE in WNL. Model is wrong or smth with software? It is BE cross over study in 3 groups.


Thanks in advance

#2 Simon Davis

Simon Davis

    Advanced Member

  • Administrators
  • 1,319 posts

Posted 13 March 2017 - 09:14 PM

Hi Reader,


 I guess you're following the European guidance;

"The terms to be used in the ANOVA model are usually sequence, subject within sequence, period and formulation. Fixed effects, rather than random effects, should be used for all terms."


The problem is that by specifying the model as all fixed effects you will often find severral of the records are not estimable - this is not an issue of Phoenix but of the model.  Better statisticians than I have debated the rationale behind this e.g. http://forum.bebac.a...sc=DESC#top6766 and may comment here.



#3 Reader



  • Members
  • Pip
  • 2 posts

Posted 14 March 2017 - 08:57 AM

Hi Simon,


Yes you are right, I try to follow. thank you for the link.



0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users