Estimating prior sd for the parameter p of a beta binomial regression

Questions
2

Hi,
I’m not sure I understand exactly what you want to do, but:

  • If you want to compute the standard deviation during sampling, you need to use theano operators, as the variables are tensors, not scalars.
  • If you want to compute it after sampling, you can just use the trace and do numpy operations on it, as the trace contains scalars.

Hope this helps :vulcan_salute:
PS: Any reason why you’re not using the built-in Beta-Binomial distribution?