Hello ALL! Some times ago I started to learn and play with Bayesian stuffs. Many advice use of WinBUGS for Bayesian inference Using Gibbs Sampler. However, WinBUGS is discontinued, and now, development is under OpenBUGS. I wasn't lazy, so I installed both and tried out. In more than 90% of cases they give comparable outcome. But in few cases I got substantial differences. Recently, I read nice things about JAGS (Just Another Gibbs Sampler). However, I did not try it. Since I am using Linux, Fedora 64-bit, and have Windows XP under VirtualBox, I even tested WinBUGS/OpenBUGS under bot Linux and Windows. Same, occasional differences repeated. My question is, as in the Subject: which BUGS should one use? Sintax is similar. They are all nicely integrated with R. But, what would be the best for a beginner? And for the future too? Best, PM
Hi Petar, I've tried to build a model with WinBugs and found out that there is a limit related to defining variables with discrete distributions. Sorry I can't remember the details but I think it did not support more than 5 discrete variables as parent of another variable or something like that. JAGS did not have the same limitation and it let me complete my work. Again, sorry about not being clear enough about my problem with WinBugs, but at least I can say that JAGS worked as expected. Maybe there was a way of handling the problem in WinBugs, but I could not find it. Kind regards Seref On Thu, Aug 30, 2012 at 1:21 PM, Petar Milin <pmilin@ff.uns.ac.rs> wrote:> Hello ALL! > Some times ago I started to learn and play with Bayesian stuffs. Many > advice use of WinBUGS for Bayesian inference Using Gibbs Sampler. > However, WinBUGS is discontinued, and now, development is under > OpenBUGS. I wasn't lazy, so I installed both and tried out. In more than > 90% of cases they give comparable outcome. But in few cases I got > substantial differences. Recently, I read nice things about JAGS (Just > Another Gibbs Sampler). However, I did not try it. > > Since I am using Linux, Fedora 64-bit, and have Windows XP under > VirtualBox, I even tested WinBUGS/OpenBUGS under bot Linux and Windows. > Same, occasional differences repeated. > > My question is, as in the Subject: which BUGS should one use? Sintax is > similar. They are all nicely integrated with R. But, what would be the > best for a beginner? And for the future too? > > Best, > PM > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >[[alternative HTML version deleted]]
Whit Armstrong is building a R-centric BUGS (RcppBugs) which looks very powerful as well: you might talk to him about it. www.rinfinance.com/agenda/2012/talk/WhitArmstrong.pdf https://github.com/armstrtw/rcppbugs Cheers, Michael On Thu, Aug 30, 2012 at 3:12 PM, Seref Arikan <serefarikan at kurumsalteknoloji.com> wrote:> Hi Petar, > I've tried to build a model with WinBugs and found out that there is a > limit related to defining variables with discrete distributions. Sorry I > can't remember the details but I think it did not support more than 5 > discrete variables as parent of another variable or something like that. > JAGS did not have the same limitation and it let me complete my work. > Again, sorry about not being clear enough about my problem with WinBugs, > but at least I can say that JAGS worked as expected. Maybe there was a way > of handling the problem in WinBugs, but I could not find it. > > Kind regards > Seref > > > On Thu, Aug 30, 2012 at 1:21 PM, Petar Milin <pmilin at ff.uns.ac.rs> wrote: > >> Hello ALL! >> Some times ago I started to learn and play with Bayesian stuffs. Many >> advice use of WinBUGS for Bayesian inference Using Gibbs Sampler. >> However, WinBUGS is discontinued, and now, development is under >> OpenBUGS. I wasn't lazy, so I installed both and tried out. In more than >> 90% of cases they give comparable outcome. But in few cases I got >> substantial differences. Recently, I read nice things about JAGS (Just >> Another Gibbs Sampler). However, I did not try it. >> >> Since I am using Linux, Fedora 64-bit, and have Windows XP under >> VirtualBox, I even tested WinBUGS/OpenBUGS under bot Linux and Windows. >> Same, occasional differences repeated. >> >> My question is, as in the Subject: which BUGS should one use? Sintax is >> similar. They are all nicely integrated with R. But, what would be the >> best for a beginner? And for the future too? >> >> Best, >> PM >> >> ______________________________________________ >> R-help at r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >> > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.