s-news
[Top] [All Lists]

Re: Help with mixed logistic model ( beta-binomial) - Correction

To: "Hunsicker, Lawrence" <lawrence-hunsicker@uiowa.edu>, <s-news@lists.biostat.wustl.edu>
Subject: Re: Help with mixed logistic model ( beta-binomial) - Correction
From: "Stella Karuri" <swkaruri@insightful.com>
Date: Fri, 28 Mar 2008 15:49:05 -0700
In-reply-to: <825A13A34899AB40A5039B9E1B71939F01DE3160@HC-MAIL12.healthcare.uiowa.edu>
References: <825A13A34899AB40A5039B9E1B71939F01DE3160@HC-MAIL12.healthcare.uiowa.edu>
Thread-index: AciPgNzAUl+Ungl+R2SBt8lVoLdBYwBpK/7g
Thread-topic: [S] Help with mixed logistic model ( beta-binomial) - Correction

fit1 <-glmmPQL(fixed = any~FIRST.SE, random = 1|PROVUSRD, )

fitVal <-fitted(fit1,levels=1)

grouped.fit <- data.frame(id=names(fitVal), fitVal)

 

or if you want the fitted values in lists with each list corresponding to a center then try

 

fit1 <-glmmPQL(fixed = any~FIRST.SE, random = 1|PROVUSRD, )

fitVal <-fitted(fit1,levels=1, asList=T)

 

 

Stella

 

 

 


From: s-news-owner@lists.biostat.wustl.edu [mailto:s-news-owner@lists.biostat.wustl.edu] On Behalf Of Hunsicker, Lawrence
Sent: Wednesday, March 26, 2008 1:35 PM
To: s-news@lists.biostat.wustl.edu
Subject: [S] Help with mixed logistic model ( beta-binomial)

 

Greetings, all:

I have run a mixed logistic model to explore the variation in center practice in ordering a diagnostic test in patients initiating hemodialysis, corrected for baseline patient level covariates (fixed effects), using glmmPQL in the MASS library. 

The call:  glmmPQL(fixed = any~FIRST.SE, random = 1|PROVUSRD, …)

The model runs successfully and I have the output.  But for graphical purposes I’d like to be able to graph the estimated likelihood of testing by center (PROVUSRD).  Is there a neat way to get from the model object a list of the fitted values for each center, rather than for each patient? 

Many thanks in advance for any help that you can give me on this.

Larry Hunsicker

<Prev in Thread] Current Thread [Next in Thread>