Then I would say what you have is at least reasonable. Two quick comments, however. First, uniform priors are often not a great idea because the discontinuities at each end of the interval give the sampler a hard time. Sampling is often easier if you use something like a gamma prior to “encourage” the sampler to stay >=0. Second, you have a large number of divergences (the tick marks along the x-axes) that should discourage you from interpreting the results you are currently getting. I suspect that these two points are related, but the divergences could arise from other sources as well.
1 Like