s-news
[Top] [All Lists]

[S] na.include

To: "s-news group" <s-news@utstat.toronto.edu>
Subject: [S] na.include
From: "WayWay M. Hlaing" <wayway@gate.net>
Date: Fri, 27 Mar 1998 08:04:07 -0500
Sender: owner-s-news@wubios.wustl.edu
Hello!
 
I am a relatively new user of Splus.  Upon reading online manuals of Nonlinear Mixed Effects Model
(NLME) & na.include, I am still not sure what exactly splus does when you include na.action="">
in the model statement.  The model statement is attached below.
 
Questions: (1)  Does Splus do the imputation automatically for all the NAs for all variables?  If so,
 with what value - mean or median? 
 
(2)  what does it mean by adding a level?  (online manual of na.include).  Many of the variables in my dataset
 are continuous variables.
 
(3) am I using the option na.include incorrectly?
 
Any help in explaining this is greatly appreciated.  I am using SPLUS4, rel.3 on winNT, analyzing a longitudinal
data set with many of the timed visits have NAs; NLME with 4-parameter GOMPERTZ function. 
 
Model Statement:
 
nlme(BMI ~ BMI.FUNC(A,B,C,D,AGE), fixed=list(A~., B~., C~., D~.),
random=list(A~., B~.),  cluster=~ID,
start=list(fixed=c(16.71, 2.08, 2.33, -0.34)), na.action="">,
control=list(tolerance=0.0001, pnls.tolerance=0.001, lme.tolerance=0.000001),
data="" verbose=T)
 
Thank you in advance, W. Hlaing, University of South Florida, Tampa, FL
 
           whlaing@com1.med.usf.edu
 
<Prev in Thread] Current Thread [Next in Thread>
  • [S] na.include, WayWay M. Hlaing <=