[R] R crashes when communicating with JAGS

[Uwe Ligges]

Works for me with JAGS 3.1.0.

Uwe Ligges


On 11.08.2011 14:09, David Wooff wrote:
> There is a thread on this topic already:
> http://finzi.psych.upenn.edu/Rhelp10/2010-August/250934.html
>
> I'm rather mystified by a similar problem and wondering whether I've
> overlooked something obvious. I'm running with latest versions of R and
> all packages, and latest version of JAGS running under Windows 7.
>
> Here's the problem. I have some source code. It's given below - standard
> stuff - seeds example from BUGS volume 1. I'm using R2WinBUGS only to
> create the model file, and R2jags as a front end to JAGS. I have the
> same problem whether or not I use R2jags.
>
> 1. If I start RGUI and source the code from a source file. R crashes
> when it tries to execute jags().
>
> Problem Event Name: APPCRASH
> Application Name: Rgui.exe
> Application Version: 2.131.56322.0
> Application Timestamp: 4e171fac
> Fault Module Name: StackHash_486a
> Fault Module Version: 6.1.7601.17514
> Fault Module Timestamp: 4ce7ba58
> Exception Code: c0000374
> Exception Offset: 000ce653
> OS Version: 6.1.7601.2.1.0.768.3
> Locale ID: 2057
> Additional Information 1: 486a
> Additional Information 2: 486a33f7a181da7ba2feae88b7ee12b6
> Additional Information 3: 9e02
> Additional Information 4: 9e02b3fab8ac2ab41c0e936dfef5c6d4
>
> 2. If I start RGUI and cut and paste the code from a source file.
> Everything works fine.
>
> 3. If, after (2) I source the code - and other examples from source
> files - everything works fine.
>
> This seems to indicate that there is some kind of initialization which
> needs to happen before JAGS will work; that this initialization is
> carried out when typing commands into the console, but not carried out
> when sourcing identical commands from a source file.
>
> I have tried delaying issuing the jags command when sourcing (using
> readline()) in case the issue was with readiness of the bugs model being
> written to file; but the crash still happens with any such delay.
>
> I'd be most grateful for tips to circumvent this behaviour.
>
> Cheers,
>
> David
>
> Here's the test code:
> #-------------------------------------------
> require(R2WinBUGS)
> require(rjags)
> require(R2jags)
> ## write model file:
> print(list.modules())
> write.model(con="blah.bug",model=function() {
> for( i in 1 : N ) {
> r[i] ~ dbin(p[i],n[i])
> b[i] ~ dnorm(0.0,tau)
> logit(p[i]) <- alpha0 + alpha1 * x1[i] + alpha2 * x2[i] +
> alpha12 * x1[i] * x2[i] + b[i]
> }
> alpha0 ~ dnorm(0.0,1.0E-6)
> alpha1 ~ dnorm(0.0,1.0E-6)
> alpha2 ~ dnorm(0.0,1.0E-6)
> alpha12 ~ dnorm(0.0,1.0E-6)
> tau ~ dgamma(0.001,0.001)
> sigma <- 1 / sqrt(tau)
> })
> g<-jags(model.file='blah.bug',
> data = list(r = c(10, 23, 23, 26, 17, 5, 53, 55, 32, 46, 10, 8, 10, 8,
> 23, 0, 3, 22, 15, 32, 3),
> n = c(39, 62, 81, 51, 39, 6, 74, 72, 51, 79, 13, 16, 30, 28, 45, 4, 12,
> 41, 30, 51, 7),
> x1 = c(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1),
> x2 = c(0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1),
> N = 21),
> parameters=c('alpha0', 'alpha1','alpha2','alpha12','tau'),
> inits=list(alpha0 = 0, alpha1 = 0, alpha2 = 0, alpha12 = 0, tau = 1) ,
> n.iter=10000)
> #-------------------------------------------
>
>