I'm modeling the probability of appendicitis in children, given their
gender, white blood cell count (wbc) and a clinical estimate of the
probability of appendicitis (priorprob) provided by the treating
clinician (doc), after taking a history and performing a physical
examination (see BUGS model below).
I would like to incorporate information from a prior study which
studied the same outcome, but evaluated only two of the current
predictors (gender and white blood cell count). What is a reasonable
approach to incorporating this prior information on a subset of the
predictors in the current model. Thanks. --Dale
# Likelihood
for (i in 1:n) {
appy[i] ~ dbin(p[i], 1)
logit(p[i]) <- b.0 + b.gender*gender[i] + b.wbc*wbc[i] +
b.priorprob*priorprob[i] + u[doc[i]]
}
# m docs
for (j in 1:m) {
u[j] ~ dnorm(0,tau)
}
# independent normal priors
b.0 ~ dnorm(0,0.0001)
b.gender ~ dnorm(0,0.0001)
b.wbc ~ dnorm(0,0.0001)
b.priorprob ~ dnorm(0, 0.0001)
# Hyperprior
tau ~ dgamma(0.001,0.001)
-------------------------------------------------------------------
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
|