I am trying to predict occurrence of a species (survey success, within a
regular grid overlaid on past survey locations) using a CAR model with
environmental covariates. My initial model was adapted from Gelfand et al.
2005. (Modelling species diversity through species level hierarchical
modelling. Royal Stat Soc C Appl. Statist. 54, Part 1, pp. 1-20) but they
use a car.normal and I would like to rewrite it as a car.proper. I have
looked over the Geobugs examples for car.proper (eg, lip cancer) but am new
at Winbugs and unsure how to adapt them to this problem. Could anyone point
me to a car.proper example similar to this problem?
More generally, what are the pros and cons of using car.normal vs.
car.proper for this type of question?
Thanks very much.
the car.normal model is:
# likelihood
for (i in 1 : N_nonzeroy) {
y[ind[i]] ~ dbin(p[ind[i]], n[ind[i]])
}
for(i in 1:N_LOC){
logit(p[i]) <- rho[i]+xbeta[i]+mu
xbeta[i]<-beta[1]*tri[i] + beta[2]*con[i]
}
# CAR prior distribution for spatial random effects:
rho[1:N_LOC] ~ car.normal(adj[], weights[], num[], tau)
for(k in 1:sumNumNeigh) {
weights[k] <- 1
}
# other priors
mu ~ dflat()
for ( i in 1:2) { beta[i] ~ dnorm(0, 1) }
vrho ~ dnorm(0, 0.2) I(0,)
tau <- 1/vrho
}
--
Carlos Carroll, Ph.D.
Klamath Center for Conservation Research
PO Box 104
Orleans, CA 95556
-------------------------------------------------------------------
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
|