I've certainly thought about it in the context of survival analysis, where the outcome consists of: A time to event variable, and A censoring indicator showing whether the failure time was censored In this case it would be useful to write the model with a bivariate outcome constructed on the fly: c(Time[i], Censored[i]) ~ dcensored.weibull(...) However, I'm not sure I want to go down this route as it breaks some fundamental assumptions the compiler is built on. I'm not saying it's impossible but...
As far as I know - and I wrote the parser - JAGS has never supported multiple var statements. I've checked all the way back to JAGS 3.0.0 to make sure I've not accidentally taken this feature out but I can find no trace of it. On the plus side, the JAGS parser ignores newlines, so you can absolutely split your single var statement over multiple lines with comments like this: # AFM11.jags # var N, # number of patients okdose[N], # a tox-free dose for each pt aedose[N], # a toxic dose for each pt mtd[N],...
Ensure that BaseRNGFactory::setSeed correctly resets the sampler factory.
'Logical node' not defined/delineated in User manual 5.0.0
For "logical" read "deterministic". I have updated the manual to reflect this change. This is a hangover from the original BUGS program which described deterministic nodes defined by the relation <- as "logical". I never liked this terminology as it implies that the value of the resulting node is Boolean. I obviously did not clean up the JAGS manual well enough. Thanks for the feedback.
Branch merge from default
Fix logical node -> deterministic node in user manual (PR#65)
Last word on this before I sign off for now. If you supply FCLIBS at configure time then this will override the macro, e.g. on Linux I can do FCLIBS="-lgfortran -lm" ./configure ... and this build JAGS without linking to libquadmath (I have omitted a lot of -L/path/to/libs flags but it still seems to work). However, you can still pick up a recursive dependency on libquadmath from a linked library. For example libflexiblas is linked to libquadmath on my Linux desktop so if I rely on flexiblas for...