1
$\begingroup$

How do I extract a credible interval ala confint on a glm object when working with the object returned by bayesglm() in arm?

$\endgroup$
1
  • $\begingroup$ You could write a method for confint for class bayesglm. $\endgroup$ Feb 20 '20 at 17:57
2
$\begingroup$

Since I do not have enough rep to comment I will put my comment as an answer. See this article, https://it.unt.edu/sites/default/files/bayesglm_jds_jan2011.pdf. Page 7 and up will be useful for you.

$\endgroup$
3
  • $\begingroup$ thanks, the use of sim() + quantile() is what i was getting at with my answer. guess this is the way to do it for now. $\endgroup$
    – user4733
    Jan 29 '15 at 20:26
  • 1
    $\begingroup$ Well it appears you can't just use sim and quantile as bayesglm isn't actually finding a posterior. For this reason I try to avoid bayesglm. You could do the fit in JAGS. JAGS output gives the credible intervals. $\endgroup$ Jan 29 '15 at 20:33
  • 1
    $\begingroup$ I'm a little confused by the terminology in the pdf. What does "’simulates’ which are used in place of an actual empirical distribution" mean? what does sim() on a bayesglm() object do, if not sample from the posterior? $\endgroup$
    – user4733
    Jan 29 '15 at 21:02
0
$\begingroup$

credible intervals can be inferred using sim() with quantile(). But I would still be interested if there's a more direct way to do this.

$\endgroup$

Your Answer

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

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