I've just started running a large MCMC analysis (multi-state model, 3 states, 27 years, 5 hyperdistributions w/ around 20 betas in each hyperdistribution)
While it is running (iterations increasing), I am repeatedly seeing the following message in the run window:
"error number 96 returned from DPOCO in MCMC" and also occasionally "error number 76 returned from DPOCO in MCMC".
Wondered if anyone knew what specifically the problem might be so that I can decide whether to scrap it or not.
Thanks for the help.
MCMC error message
2 posts
• Page 1 of 1
MCMC error message
by gillian » Mon Sep 10, 2007 4:55 pm
- gillian
- Posts: 4
- Joined: Thu Apr 22, 2004 4:32 pm
- Location: Lakeview, MT
MCMC error message
by gwhite » Mon Sep 10, 2007 5:18 pm
Gillian:
The error message is coming from the routine inverting the variance-covariance matrix that you've specified. The VC matrix is singular, which suggests that you haven't specified the matrix correctly. I would have to see the structure of the matrix to (maybe) tell you what might be wrong.
Gary
The error message is coming from the routine inverting the variance-covariance matrix that you've specified. The VC matrix is singular, which suggests that you haven't specified the matrix correctly. I would have to see the structure of the matrix to (maybe) tell you what might be wrong.
Gary
- gwhite
- Posts: 340
- Joined: Fri May 16, 2003 9:05 am
2 posts
• Page 1 of 1
Who is online
Users browsing this forum: No registered users and 2 guests