search

gertvv / gemtc

GeMTC R package: model generation for network meta-analysis
GNU General Public License v3.0
44 stars 25 forks source link

Relative effect models with more than 3-arm trials won't work in BUGS #28

Closed gertvv closed 9 years ago

gertvv commented 10 years ago

I'm getting this error:

NIL dereference (read)

 GraphNodes.AddParent   [0000019FH] 
    .list   GraphNodes.List NIL
    .parent GraphNodes.Node NIL

Which, according to the manual:

d) 'NIL dereference (read)' can occur at compilation in some circumstances when an inappropriate transformation is made, for example an array into a scalar.

It seems to have to do with the fact that I'm subsetting two dimensions of the array at once in the multi-arm code:

Omega[i,1:(na[i]-1),1:(na[i]-1)] <- inverse(Sigma[i,1:(na[i]-1),1:(na[i]-1)])
m[i,2:na[i]] ~ dmnorm(delta[i,2:na[i]],Omega[i,1:(na[i]-1),1:(na[i]-1)])

I'm not sure whether there is a good workaround for this. The model works fine in JAGS. This might just convince me to drop support for BUGS.

gertvv commented 9 years ago

Closing because support for BUGS has been dropped (e4f1409b01).