I tried mcmc_parcoord and nothing really stood out, even after normalizing the parameters to put them on the same scale. One catch is that I have about 70 parameters, mostly due to ABn_indiv
and unit_ABn
, so a lot of plots are not very useful, especially the pairs plots. I tried using the stan_checkdivergences.R function at Concentration of divergences thread, but that doesn’t work with saved stanfit objects, due to an issue discussed on the old Google Group for Stan.
I was hoping that I had done something that would stand out to someone as obviously flawed, but it looks like things won’t be that easy.