I wrote a case study on the Nearest neighbor Gaussian process (NNGP) based model and would be happy to get feedback. It would be great if someone can help me to put it on the Stan website.
Nearest Neighbor Gaussian Processes (NNGP) based models in Stan
Nearest Neighbor Gaussian Processes (NNGP) based models is a family of highly scalable Gaussian processes based models. In brief, NNGP extends the Vecchia’s approximation to a process using conditional independence given information from neighboring locations. In this case study, I will briefly review response and random-effects NNGP models, illustrate how to code NNGP in Stan, and provide examples of implementing NNGP based models in Stan.
The Rmarkdown file and the compiled HTML file are on my Github:
And here is a pdf for easier viewing. The format in the pdf is a little bit off-shape.
nngp_stan.pdf (1.6 MB)
Thanks, and wish all of you a happy Thanksgiving!