Thanks for the quick reply!
I was viewing the github repo named ‘aloctavodia’ that I linked to in my first post in this thread. I was not aware that the ‘Statistical Rethinking’ Notebooks have moved to another github repo.
Please review your very strict spam-filter settings in this forum, because it makes it difficult to post good and precise questions when the spam-filter blocks internet links so easily.
I don’t know if you are interested in feedback about the LKJ tutorial, but here goes: It requires that the reader already knows alot about all this. This is really not a tutorial for beginners. I have watched McElreath’s youtube lectures and I still consider myself a beginner who sort-of understands the main ideas. But I found your LKJ tutorial really confusing. There also seems to be remnants of the old syntax, e.g. cells 5-7 which still uses packed_L and pm.expand_packed_triangular to calculate something, but apparently its results aren’t used anywhere. I think your LKJ tutorial is really only suitable for people who are already very experienced in all this. For a beginner it is super-confusing.
I appreciate your tremendous effort already and that you probably have a million other things to do, but if you consider this topic to be important, then it may be worth your time to write a more beginner-friendly tutorial on how to do Multi-Variate Normals in pymc3.
Thanks again for all your efforts!
PS: I bought a cool hat like you have in your profile picture, expecting that it would make it easier to understand all this Bayes stuff, but pymc3 tells me there’s an 89% probability that the hat didn’t really help me, although I’m not quite sure that my model is correct.