I'm repeatedly getting almost no mixing of the walkers, see the attached figure. In other examples, I saw that walkers didn't move at all (which seems to contradict the statement of unit-acceptance). In all these runs, mu gets arbitrarily close to 0, before the patience setting kicks in. The posterior distribution is unimodal and has some mildly bent degeneracies but nothing too crazy. Do you have any advice?
I'm repeatedly getting almost no mixing of the walkers, see the attached figure. In other examples, I saw that walkers didn't move at all (which seems to contradict the statement of unit-acceptance). In all these runs, mu gets arbitrarily close to 0, before the
patiencesetting kicks in. The posterior distribution is unimodal and has some mildly bent degeneracies but nothing too crazy. Do you have any advice?