CosmoMC: some fault in proposal matrix

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
lele zhang
Posts: 4
Joined: November 02 2006
Affiliation: shandong university

CosmoMC: some fault in proposal matrix

Post by lele zhang » January 26 2007

Hi,
There are some faults when cosmomc evaluate the proposal matrix and then
the chains was cut off. I only added to a new parameter. I don't know why
Proposal matrix has negative or zero eigenvalues. what's matter in this condition?
and how to avoid. thanks

Chain 3 MPI communicating
Chain 1 MPI communicating
Current convergence R-1 = 0.2775182 chain steps = 2258
Proposal matrix has negative or zero eigenvalues
Proposal matrix has negative or zero eigenvalues
Proposal matrix has negative or zero eigenvalues

Pascal Vaudrevange
Posts: 50
Joined: March 26 2006
Affiliation: DESY
Contact:

CosmoMC: some fault in proposal matrix

Post by Pascal Vaudrevange » February 06 2007

I ran into negative eigenvalues of the proposal matrix when using a primordial powerspectrum that is NaN for some values of k.
What is this parameter that you added? Some more information would be useful.

Post Reply