I am doing some Bayesian analysis in OpenBUGS and I need to specify a joint prior distribution for parameters $\gamma_1$ and $\gamma_2$ (the Jeffreys prior related to my model), such that $$\pi(\gamma_1,\gamma_2)\propto \sqrt{\frac{1}{(1-\gamma_1)(1-\gamma_1-\gamma_2)}},$$ subject to restrictions $\gamma_1\in (0,1)$ and $\gamma_2\in (0,1-\gamma_1)$.

I tried to consider the 'zeros trick' but it has appeared some problems within log functions due to the use of dflat() prior to $\gamma_1$ and $\gamma_2$.

May someone please help with the implementation of this kind of restricted distribution in OpenBUGS? Is it possible to implement such a distribution?


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.