I have the following

$$x\sim NegativeBinomial(r,\theta)$$

$$\theta \sim beta(a,b)$$

and proved that posterior is

$$\theta|x \sim beta(\sum x_{i} +a,nr+b)$$

I would like to find the posterior predictive distribution so I calculate $$f(y|x)=\int_{0}^{1} f(y|\theta)p(\theta|x)d\theta$$ which is equal to $$\begin{pmatrix} y+r-1\\ y \end{pmatrix}\frac{B(\sum x_{i}+a+y,r+nr+b)}{B(\sum x_{i}+a,nr+b)}$$

and I don't truly know which distribution is this , but it seems similar to Beta-Binomial.

Any idea or help would be great.


1 Answer 1


There is a distribution called a "Beta Negative Binomial" distribution, which properties are described on Wikipedia, which corresponds to this case. Its main properties are described on this page, but I wonder how useful it is to give it a name.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.