how to plot the results of a GLMM

From: Rebecca Doyle <asremlforum_at_VSNI.CO.UK>
Date: Mon, 18 May 2009 06:07:01 +0100

Dear Users,

I have performed a GLMM (binary, logit) in ASReml to analyse the difference in response between two treatment groups to several related situations (it is a behavioural study). I have also used a simple PREDICT statement to give me values for each treatment for one specific situation. I am now trying to graph the response (Y) according to situation (X) for the two treatment groups.

The problem I have is that when I use the estimates in my .asr file to plot the line the results I get are different to the predicted values for that same situation and generally the points for the lines are much lower than I thought they should be. The equations I am using to generate my Y values for the lines are as follows:

Z = mu estimate + (situation estimate * situation number) + (treatment estimate * treatment number)

From there:
Y = 1/(1+e(-Z))

If you can see where I am going wrong or have another kind of suggestion I would be very appreciative.
Rebecca

-------------------- m2f --------------------

Sent using Mail2Forum (http://www.mail2forum.com).

Read this topic online here:
http://www.vsni.co.uk/forum/viewtopic.php?p=543#543

-------------------- m2f --------------------
Received on Tue May 18 2009 - 06:07:01 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)