s-news
[Top] [All Lists]

Re: [S] LME - fixed effects model and missing values

To: s-news@wubios.wustl.edu
Subject: Re: [S] LME - fixed effects model and missing values
From: ziv shkedy <ziv.shkedy@luc.ac.be>
Date: Thu, 21 Jan 1999 16:12:45 +0100 (MET)
Sender: owner-s-news@wubios.wustl.edu
>Date: Wed, 20 Jan 1999 11:21:48 -0500
>From: Jose Pinheiro <jcp@research.bell-labs.com>
>To: ziv.shkedy@luc.ac.be
>CC: bates@stat.wisc.edu
>Subject: Re: [S] LME - fixed effects model and missing values
>
>I am sorry, but I am still a bit confused with your description. What
>I understand is that there are some missing responses for some of the
>subjects. What na.omit will do is remove only the observations that
>are missing, not all observations on individuals that have at least
>one missing value. The estimation in lme is also likelihood based, so
>it uses all available information. One thing to notice is that the
>structure of the data for an lme analysis should be like
>
>  Subject time response
>    1      1    0.1
>    1      2    0.4
>    1      3     NA
>    1      4    0.9
>    2      1    0.05 
>    2      2     NA
>    2      3    0.7
>    2      4    1.0
>
>and not one line per subject. na.omit would convert the above dataset
>to
>
>  Subject time response
>    1      1    0.1
>    1      2    0.4
>    1      4    0.9
>    2      1    0.05 
>    2      3    0.7
>    2      4    1.0
>
>and this is naturally handled in lme. If there are NAs in other
>variables in the data at other times in which the response is
>observed, then they will also be removed by na.omit, as the function
>can not differentiate between the variables. In this case, you should
>remove the variables that are not used in the analyis and contain NAs,
>before calling lme. 
>
>I am not sure if I am answering what you asked, so let me know if I
>missed something.
>
>Best,
>
>--Jose'
>
>-----------------------------------------------------------------------------
> Jose' Pinheiro
> Bell Laboratories                     jcp@research.bell-labs.com
> 600 Mountain Avenue, Room 2C-258      office: (908) 582-2390
> Murray Hill, NJ 07974                 fax:    (908) 582-3340
>-----------------------------------------------------------------------------
>
>
=========================================================
Ziv Shkedy
Biostatistics                   
Center for Statistics           
Limburgs Universitair Centrum
Universitaire Campus, department WNI
B-3590 Diepenbeek, Belgium
Tel: +32-(0)11-26.82.57
e-mail: ziv.shkedy@luc.ac.be 
=========================================================

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu.  To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message:  unsubscribe s-news

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