In one of my MultiNest runs four modes were found, and the maximum likelihood and MAP parameter vector was reported to be exactly the same in all of them. Could there be an issue with the numbers printed in the stats file, or is it normal? None of the local MAP vectors matches the global MAP vector that is reported in the summary file. In my all other runs (with very similar models) maximum likelihood and MAP parameters are never exactly the same, and the global MAP vector is just one of the mode-specific MAP vectors.
Here the stats and summary files for this run:
run1_nlive10k_eff0p3_noCONST_MMon_noIS_tol-1stats.txt
run1_nlive10k_eff0p3_noCONST_MMon_noIS_tol-1summary.txt
In one of my MultiNest runs four modes were found, and the maximum likelihood and MAP parameter vector was reported to be exactly the same in all of them. Could there be an issue with the numbers printed in the stats file, or is it normal? None of the local MAP vectors matches the global MAP vector that is reported in the summary file. In my all other runs (with very similar models) maximum likelihood and MAP parameters are never exactly the same, and the global MAP vector is just one of the mode-specific MAP vectors.
Here the stats and summary files for this run:
run1_nlive10k_eff0p3_noCONST_MMon_noIS_tol-1stats.txt
run1_nlive10k_eff0p3_noCONST_MMon_noIS_tol-1summary.txt