This is a big piece of work I have been doing with Sudipto Banerjee and my then PhD advisor Benoît Liquet. The work is focused on statistical modeling of geographic data, where there is some heterogeneity (nonstationarity) in the behavior of the interest variable. For those who are brave enough, you can dive into the paper, which is still in revision. There is a lot in it for people who are interested in hierarchical modeling, in particular concerning Monte-Carlo strategies. However, if you are more interested in case studies, you can go to the GitHub page, then download a R package to install on your machine, and a vignette that will give you a practical intuition of the model and a step-by-step guidance on how to use it.
If you have questions, suggestions, remarks, if you want explanations, if something is unclear, if you want some help with the package, if you want me to come to your lab and talk about it (I can dream), just email me!