Hello All,

I am estimating a Latent Variable Model, of the sort Y[i,j] = b0[j]+b1[j]Z[i], where the Y's are observed and the Z's are not.  The data are counts of sentences in party manifestoes over the past 50 years.  The Z's are distributed normally with a party-specific mean for the first election and a random walk prior thereafter.  Now, I would like to take the Z values and use them as the response variable in a regression, but I don't want the predictors in this second stage to influence the value, so I would need to use something like the cut function.  So, here's the basic setup:

Y[i,j] ~ dnorm(mu.y[i,j], tau[j])
    mu.y[i,j] <- b[j,1] + b[j,2]*Z[i]


Z.cut[i] <- cut(Z[i])

Z.cut[i] ~ dnorm(mu.z[i], tau)
     mu.z[i] <- gamma[1] + gamma[2]* X[i,1] + gamma[3]*X[i,2]

My problem is, that WinBUGS doesn't seem to want to let me say Z.cut[i] <- cut(Z[i]) AND Z.cut[i] ~ dnorm( , ).  Does anyone know how I can get around this problem?

Thanks for your help,


Dave Armstrong
University of Maryland
Dept of Government and Politics
3140 Tydings Hall
College Park, MD 20742
Office: 2103L Cole Field House
Phone: 301-405-9735
e-mail: [log in to unmask]
web: www.davearmstrong-ps.com

Facts are meaningless.  You can use facts to prove anything that's even remotely true. - Homer Simpson ------------------------------------------------------------------- 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