Measurement error in a predictor that is a proportion

[jd_c]

Thanks! I think I implement something similar with beta_proportion.

@LucC

P.s.: i see that the use of a beta distribution was already suggested. There is no need though to put bounds very close to zero and one. You should be fine putting them at zero and one directly

I’m not sure what you mean here. The Stan documentation says theta can’t be zero or one, so that is why I added a little bit to proportions that are zero and subtract a little from those that are one.

Would you have any idea how to add a varying effect such as described in this post? In my real application, I have many measurements per grouping level, and I would like information about true_proportion to be correlated within groups.