Is there a prior that's commonly used for "index" or "score" type variables that are user-defined as a weighted sum of a small number of variables (sometimes with pre-defined interaction contributions)?

I know the gamma, inverse gamma, binomial, negative binomial, poisson, etc distributions are strictly positive, but they correspond to processes that don't describe something like an index/score type variables well. Distributions of index/score type of values are usually overdispersed relative to these distributions.

I could construct my own mixture distribution, but I was wondering if there is are standard practices for this kind of data.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.