expected variable name error pos 98349 WInBUGS in R

I am using a random intercept model with SITEID as random and NAUSEA as
outcome.

I tried using a dataset without missing values and changed my model
statement accordingly but still get the same error. Follwoing in an excerpt.

anal.data <- read.table("nausea.txt", header=T, sep="\t")
list(names(anal.data))

[[1]]
[1] "SITEID" "NAUSEA"

#anal.data <- read.csv("simuldat.csv", header=T)
attach(anal.data)

        The following object(s) are masked from anal.data ( position 3 ) :
         NAUSEA SITEID

        The following object(s) are masked from anal.data ( position 4 ) :
         NAUSEA SITEID

data.bugs <-

list("NAUSEA"=NAUSEA,"SITEID"=SITEID,"n.samples"=n.samples,"n.sites"=n.sites,"n.params"=n.params)

bugsData(data.bugs, fileName = "nauseadata.txt")
inits.bugs <- list("alpha"=rep(0,n.sites), "tau"=1)
bugsInits(list(inits.bugs), fileName = "nauseainit.txt")
modelCheck("nausea_random.txt")       # check model file

model is syntactically correct

modelData("nauseadata.txt")                     # read data file

expected variable name error pos 98349
*MODEL*
model
{
for (i in 1:n.samples)
{NAUSEA[i] ~ dbin(p[i],1)
logit(p[i]) <- alpha[SITEID[i]]}
#for(k in 1:n.params)
#{b[k]~ dnorm(0.0,tau)}
for (j in 1:n.sites)
{alpha[j]~dnorm(0.0,1.0E-10)}

tau ~ dgamma(0.001,0.001)
}
*Dataset:*
SITEID NAUSEA
1 0
1 1
1 1
1 0
1 1
1 1
1 0
1 1
1 1
1 1
1 0
1 1
1 0
1 1
1 1
1 1
1 0
1 0
1 0
1 1

-Anamika

	[[alternative HTML version deleted]]

I did run the same model in WinBUGS, letting WinBUGS generate initials but
now it gives me the error
'data value already given to this quantitiy' using a sample dataset and the
cursor blinks in the highlighted section.

list(NAUSEA=c(1,0,0,0,0,1,1,1,0,1,1,0),SITEID=c(1,1,1,1,1,1,2,2,2,2,2,2),
n.sites=2, n.samples=12)

Any thoughts? Whats the listserv for WinBUGS error?

Thanks
Anamika
On Tue, Dec 9, 2008 at 5:19 AM, Uwe Ligges
<ligges at statistik.tu-dortmund.de>wrote:

Hard to help without seeing the actual files.
Anyway, this is a BUGS question rather than  an R question, because the
error comes from the BUGS interpreter. Hence wrong mailing list.

Best wishes,
Uwe Ligges




Anamika Chaudhuri wrote:

  I am using a random intercept model with SITEID as random and NAUSEA as

outcome.

I tried using a dataset without missing values and changed my model
statement accordingly but still get the same error. Follwoing in an
excerpt.

anal.data <- read.table("nausea.txt", header=T, sep="\t")
list(names(anal.data))

[[1]]
[1] "SITEID" "NAUSEA"

#anal.data <- read.csv("simuldat.csv", header=T)
attach(anal.data)

       The following object(s) are masked from anal.data ( position 3 ) :
        NAUSEA SITEID

       The following object(s) are masked from anal.data ( position 4 ) :
        NAUSEA SITEID

data.bugs <-

list("NAUSEA"=NAUSEA,"SITEID"=SITEID,"n.samples"=n.samples,"n.sites"=n.sites,"n.params"=n.params)

bugsData(data.bugs, fileName = "nauseadata.txt")
inits.bugs <- list("alpha"=rep(0,n.sites), "tau"=1)
bugsInits(list(inits.bugs), fileName = "nauseainit.txt")
modelCheck("nausea_random.txt")       # check model file

model is syntactically correct

modelData("nauseadata.txt")                     # read data file

expected variable name error pos 98349
*MODEL*
model
{
for (i in 1:n.samples)
{NAUSEA[i] ~ dbin(p[i],1)
logit(p[i]) <- alpha[SITEID[i]]}
#for(k in 1:n.params)
#{b[k]~ dnorm(0.0,tau)}
for (j in 1:n.sites)
{alpha[j]~dnorm(0.0,1.0E-10)}

tau ~ dgamma(0.001,0.001)
}
*Dataset:*
SITEID NAUSEA
1 0
1 1
1 1
1 0
1 1
1 1
1 0
1 1
1 1
1 1
1 0
1 1
1 0
1 1
1 1
1 1
1 0
1 0
1 0
1 1

-Anamika

       [[alternative HTML version deleted]]