1
$\begingroup$

I'd like to use gelman.diag for an MCMC chain I ran in JAGS. It is very large, so I can't provide it.

The chain contains several MVN distributions, and I use a wishart prior on the precision matrix.

Just wondering what potential sources of this error might be?

Thanks you

$\endgroup$
  • 1
    $\begingroup$ Your question is missing a crucial detail: what error? $\endgroup$ – whuber Mar 18 '13 at 19:47
  • 1
    $\begingroup$ Sorry. "Error in chol.default(W): the leading minor of order 4 is not positive definite." I have double checked, no monitors on constant nodes. $\endgroup$ – user1375871 Mar 19 '13 at 1:38
2
$\begingroup$

You may have set a monitor on a constant parameter, which will cause chol.default to choke.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.