Print

Print



Dear BUGS users,

I’m confused about specifying the initial terms in a directed second order random walk.  The specification below follows the Ice example in the BUGS 0.5 Examples Vol 2 (ver. ii):

prec1<-0.000001*tau
b[1]~dnorm(0,prec1)
b[2]~dnorm(0,prec1)
for (i in 3:N) {
        mean[i]<-2*b[i-1]-b[i-2]
        b[i]~dnorm(mean[i],tau)
}
tau~dgamma(0.01,0.01)

The tau in prec1 is explained as being necessary to provide the appropriate likelihood for tau.

But I’ve also seen examples where the initial values were specified as just:
b[1]~dnorm(0,0.001)
b[2]~dnorm(0,0.001)
omitting the tau term entirely.

And this suggests yet a third specification:
b[1]~dflat()
b[2]~dflat()

I like this third specification because it maintains compatibility with including spatial effects via car.normal, which states a dflat intercept MUST be specified along with car.normal.
I’ve tried this and it seems to “work”, but is it appropriate?

Thanks in advance for any comments.

Dennis


Dennis Heisey
Research Biologist
USGS National Wildlife Health Center


       

------------------------------------------------------------------- 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