search

lauken13 / mrpkit

Tools and tutorials for multi-level regression and post-stratification of survey data
Other
10 stars 0 forks source link

Need to test custom modeling functions #144

Closed lauken13 closed 1 year ago

lauken13 commented 1 year ago
jgabry commented 1 year ago

Right now we're requiring a formula argument to the modeling function, which will limit the custom options. We can look into relaxing that requirement.

jgabry commented 1 year ago

Actually, just noticed we have an issue open related to my comment above: #105

lauken13 commented 1 year ago

I think we will need to alway assume a formula argument because we intrinsically always need to know what variables are the outcome and which are the predictors? Do you have an example of a non-formula based algorithm so we can think it through?

jgabry commented 1 year ago

Do you have an example of a non-formula based algorithm so we can think it through?

How about a function that calls RStan or CmdStanR to run a Stan program but not via rstanarm or brms?

lauken13 commented 1 year ago

I am closing this issue for duplication but have mentioned it over in the other issue