rstanarm icon indicating copy to clipboard operation
rstanarm copied to clipboard
verified publisher icon stan-dev

Add group-level predictor(s) to MRP case study

Open jgabry opened this issue 5 years ago • 2 comments

Just a reminder for me and @lauken13 (or anyone who wants to help) that we should add a group-level predictor to the MRP case study to make it a bit more realistic.

jgabry avatar May 18 '20 20:05 jgabry

By "group-level predictor", do you mean varying slopes? If yes, then male might be modeled as having varying slopes across age, see my second question in #437.

fweber144 avatar May 28 '20 11:05 fweber144

Oops, I think I meant to be more specific and say "state-level" instead of just "group-level". Usually for MRP we would include a predictor measured at the state-level (in addition to the varying intercepts by state). Since this is simulated data I guess we can decide what variable to generate and use. Tagging @lauken13 since she's the primary author on this. Lauren what do you think?

jgabry avatar May 28 '20 16:05 jgabry