s-news
[Top] [All Lists]

One step forward two steps back

To: s-news@lists.biostat.wustl.edu
Subject: One step forward two steps back
From: "paul hughes" <u7n96@hotmail.com>
Date: Wed, 28 Feb 2001 23:06:24 -0000
Hi all,

I am getting to grips with the more basic programming of the s language(at last I here you all say). I have however come across a problem that I am unable to solve. This is it.

I import a dataset into a data frame:

import.data(FileName="C:\\Project Work\\armd.xls",
FileType="EXCEL",DataFrame="real.data")

dim(real.data)
[1] 812  22

Then I abstract the information I am interested in. The data set I have is form a longitudinal study on age related macular degeneration.

vec <- real.data$VISIT

tim <- as.vector(DATA[,3][vec==3])
length(tim)

zero<-as.vector(rep("0",203))

stim <-as.vector(DATA[,3][vec>3])

tims<-as.vector(c(zero,stim))

ti<-as.vector(rep(tim,4))

trt <-as.vector(real.data$RX)

y<- as.vector(real.data$BL)

pa.no<-real.data$ID

This set of commands gives a set of vectors that I want to use to in lme:

armd.fit2<-lme(fixed=y~ti+tims+trt,cluster=~pa.no,
data=real.data,na.action = na.omit)

If I run the above S gives an error saying it cannot allocate memory as object size is too big So I run:

options(object.size=Inf)

and then run:

armd.fit2<-lme(fixed=y~ti+tims+trt,cluster=~pa.no,
data=real.data,na.action = na.omit)

again. S stops responding and the harddisk goes nuts. Does anyone else ever have this problem or am I again at the forefront of being rubbish?

Please try an help me!

Paul H




_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.


<Prev in Thread] Current Thread [Next in Thread>
  • One step forward two steps back, paul hughes <=