My dataset contains a longitudinal process (blood pressure measurements) and survival information (time to first stroke). The standard deviation of each individual's blood pressure measurements predicts stroke risk. I try to use a joint model to fit this - with a mixed effects model for the blood pressure, giving each person their own error precision, tau, and then using 1/sqrt(tau) as a predictor in a weibull model for survival times. WinBUGS will calculate DIC nicely if I use tau in the survival model, but not if I try to transform it (I can add a constant but either inverting or square-rooting cause problems). Are there any neat tricks to get around this? On a related note, putting bounds on tau (so that I don't get zero variance for someone's blood pressure) using tailed.tau[j] <- min(tau[j], 10000) causes the same problem. Is there a workaround? ------------------------------------------------------------------- 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