Beginner Advice

@v4gadkari Of course, my modeling is never gonna get as advanced as how people make it in stan

You really don’t know that!

I would recommend that you consider starting with RStan rather than brms or any interface (or maybe better, use them both side by side). Its not that difficult to learn to use Stan, if you are highly motivated to learn (which you clearly are). But I think there are advantages to working with the Stan code yourself, in terms of beginning to grasp the concepts at a higher level and learning to be creative. This would be the “slow” route. (@jsocolar just said everything I wanted to say and much more that I could have, while I was writing this.)

As someone with a couple liberal arts degrees, I’m still surprised by the lowly place of theory in courses, books, and discussions on modeling. There’s so little. In my experience, intro courses are all about introducing students to debates about fundamental ideas and the history of the field, even though the ideas are often over their heads. There are lots of interesting and difficult debates about probability theory that, in my view, would be worth your while to start looking at. Trying to dig into the opening chapter of Cox’s Algebra of Probable Inference, for instance, could be fun

See this discussion too:

2 Likes