Due to responses to my last post, I am adding some details and
discussion. Suppose you have a simple linear model
model{
for (i in 1:10){
y[i] ~ dnorm(mu[i],tau)
mu[i] <- a + b*x[i]
}
a ~ dnorm(0,1.0E-6)
b ~ dnorm(0,1.0E-6)
tau ~ dgamma(.001,.001)
}
But some of your data is missing
x[] y[]
0 0
NA 1
2 NA
NA 3
4 NA
NA 5
6 NA
NA 7
8 NA
NA 9
END
The missing data in the y nodes is no problem, but various users have
written about how to model the missing data in the covariate x. The
missing data in the covariate can be modeled by making x[i] into a
variable node with vague parameter priors.
x[i] ~ dnorm(mu.x,tau.x)
mu.x ~ dnorm(0,1.0E-6)
tau.x ~ dgamma(.001,.001)
With luck, your missing data will be a small part of your data set, and
it won't matter whether you do all this, or just drop your missing data
from the analysis. One might imagine situations however (like the
above) in which the data from incomplete data pairs can have a
significant impact on the resulting parameter estimates. So in general
it is probably a good idea to try to make use of your fragmentary data
and attempt to model your missing data.
Finn
-------------------------------------------------------------------
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
