Skip to content

Random effect model

2 messages · Hathaikan Chootrakool, Spencer Graves

Original
Hathaikan Chootrakool
# 
Dear R-help group,

I would like to model directly following random effect model:

  Y_ik = M_ik +  E_ik  where M_ik  ~ N(Mew_k,tau_k^2)
                             E_ik  ~ N(0,s_ik^2)
  i = number of study
  k = number of treatment
---------------------------------------------------------------------------

I have practiced using the command  from 'Mixed -Effects models in S and
S-plus' as follow

fm1logit.lme <- lme(logitp~1, data=logit, random = ~1|factor(Tr))

It can be written in this model

Y_ik = Mew + B_i + E_ik  where M_i ~ N(0,sigma_b^2)
                               E_ik ~ N(0,sigma^2)


 but it is not the same what my model is.


Could somebody please point me in the right direction ?

 Sorry if this turns out to be an extreamly simple question, I'm a
 new user to R.

 Thank you very much,

 Ae
4 days later
Spencer Graves
# 
I have not seen a reply to this, so I will offer a comment.

	  1.  You are doing something very wise in reading Pinheiro and Bates 
(2000) Mixed-Effects Models in S and S-Plus.

	  2.  Unfortunately, beyond this, I don't really understand the 
question;  if others had similar difficulties, it might explain why I 
have not seen a reply.

	  3.  There is a famous book by Polya on "How to Solve It".  I regard 
the "posting guide" (www.R-project.org/posting-guide.html) as similar to 
Polya's famous book:  The posting guide can help you prepare a question 
for this group in such a way that will increase the chances of receiving 
a prompt reply.  One important suggestion is to prepare a self-contained 
toy example that someone else can copy from an email into R and test a 
few ideas in a minute or so.  Often, people find answers to their own 
questions while working through the posting guide.  Failing that, the 
resulting questions are often much easier for someone else to understand.

	  4.  Have you tried the "RSiteSearch" function?  Many questions have 
been asked and answered about lme, and this function might help you find 
what you want in a response to a previous question.

	  Good Luck.
	  spencer graves
Hathaikan Chootrakool wrote: