Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Bayesian algorithm(s) used by JASP

From the source code on GitHub, it seems like JASP uses the MCMC algorithm JAGS. Are there any other algorithms used for the Bayesian methods? In the source code on GitHub, the R package "StanHeaders" also occurs. Does that mean that Stan is used, too? By the way, "StanHeaders" is not listed in the R package list (https://jasp-stats.org/r-package-list/).

Thanks in advance!

Comments

  • Hi TJU,

    The next version of JASP includes a JAGS module. For most other Bayesian analyses, we use analytic expressions (whenever we can), BAS, or BayesFactor (two R packages). These use their own sampling methods, but that's often (not always) not MCMC. We are developing some other methods based on precompiled Stan code; in the future we hope to add a Stan module (similar to the JAGS module).

    Cheers,

    E.J.

  • Thanks!

Sign In or Register to comment.