Hi, I want to fit predicted proportions to observed ones using a dirichlet with an added multiplier allowing to control the dispersion. However, I get a "slicer stuck at value with infinite density" error. I found the suggestion of @martyn_plummer to add an offset (https://sourceforge.net/p/mcmc-jags/discussion/610037/thread/c21ef62a/) but it did not solve my problem. I found a way to make it work (see below), and for me, it seems to be quite similar to fitting the dirichlet directly (as the precision...
Hi, I want to fit predicted proportions to observed ones using a dirichlet with an added multiplier allowing to control the dispersion. However, I get a "slicer stuck at value with infinite density" error. I found the suggestion of @martyn_plummer to add an offset (https://sourceforge.net/p/mcmc-jags/discussion/610037/thread/c21ef62a/) but it did not solve my problem. I found a way to make it work (see below), and for me, it seems to be quite similar to fitting the dirichlet directly (as the precision...
Hi, The distribution of the sum of independent random variables is the convolution of their distributions. I can easily do the sum of several distributions in JAGS, however, how can I obtain the CDF of this distribution and sample from it to fit to observed data ? Note: my interest is more on the sum of weibull or gamma distributions Sorry from cross posting this question as I have asked something similar in the "open discussion" part. But I guess this is more an "help" question than a subject of...
Hi, The distribution of the sum of independent random variables is the convolution of their distributions. But how can I code a convolution in JAGS ? For example, how can I obtain the distribution of the sum of two weibull distributions ? In fact, my interest is in the CDF of the resulting distribution and I want to fit observed data to this CDF. Thanks
Hi, The distribution of the sum of independent random variables is the convolution of their distributions. But how can I code a convolution in JAGS ? For example, how can I obtain the distribution of the sum of two weibull distributions ? Thanks
Dears JAGS users, Sorry to insist on this subject, but I hope finding a way to solve...
Is there a way to send the update information to a cross platform external progress...
Hi Martyn, Long time since I tried to use your suggestion. Here is a slightly modified...