Hello
I am trying to fit a model where I use the following:
for (j in 1:2)
{
k[mont,j]~dunif(0.005, 0.5)
nu[mont,j]~dexp(k[mont,j]) I(3,)
}
I am getting a message that says " Can not sample from interval censored
gamma full conditional"
Any remedy to this ?
Kind regards
K. Georges Saikali, Ph.D., Exec. MBA
Director, Biostatistics
Forest Research Institute
(201) 427-8223
-----Original Message-----
From: (The BUGS software mailing list) [mailto:[log in to unmask]]
Sent: Friday, September 16, 2005 2:31 PM
To: [log in to unmask]
Subject: [BUGS] strange deviances in logistic model
Hello all,
We are trying to compare various multilevel logistic regression models
to
explain presence/absence of different fish species based on predictors
at
multiple scales. The problem is, with some models the deviance scores
do
not appear credible. Specifically, when adding certain predictor
variables the deviance increases, rather than decreases. This happens
even though examination of the credible interval for the parameter in
question shows that the predictor is strong and clearly non-zero. The
problem disappears in a non-hierarchical formulation of the model, when
the random effects are removed. I should also note that the problem
only
occurs for species with very few positive occurrences.
Does anyone else have examples of such a problem? Is this somehow a
side
effect of overfitting the data? Without credible deviances we are
unable
to use an information criterion, which makes model selection trickier.
I have tried reformulating the models in various ways but nothing helps.
Here is one version-- the full version also incorporates incomplete
detectability, but the problem is the same with or without that added
complication.
# Level 1. Site
for (i in 1:232) {
A[i,51] ~ dbern(P[i])
logit(P[i]) <- a[1]*A[i,10] + a[2]*A[i,32] + L2[A[i,8]] + err1[i]
err1[i] ~ dnorm(0, tau[1]) }
# Level 2. Reach-level.
for (j in 1 : 204) {
L2[j] ~ dnorm(mu1[j], tau[2])
mu1[j] <- L3[B[j,2]] }
# Level 3. Trib system.
for ( k in 1 : 24) {
L3[k] ~ dnorm(mu2[k], tau[3])
mu2[k]<- c0 + c1*C[k,2] }
Thanks for any suggestions.
Seth Wenger
Institute of Ecology
University of Georgia
Athens, GA USA
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail
[log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at
www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask] If
this fails, mail [log in to unmask], NOT the whole list
This e-mail and its attachments may contain Forest Laboratories, Inc. proprietary information that is privileged, confidential or subject to copyright belonging to Forest Laboratories, Inc. This e-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this e-mail, or the employee or agent responsible for delivering this e-mail to the intended recipient, you are hereby notified that any dissemination, distribution, copying or action taken in relation to the contents of and attachments to this e-mail is strictly prohibited and may be unlawful. If you have received this e-mail in error, please notify the sender immediately and permanently delete the original and any copy of this e-mail and any printout.
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|