search

danheck / metaBMA

Bayesian Model Averaging for Random and Fixed Effects Meta-Analysis
https://danheck.github.io/metaBMA/
28 stars 2 forks source link

changing the prior default to `cauchy(0, 0.707)` #9

Closed IndrajeetPatil closed 3 years ago

IndrajeetPatil commented 4 years ago

Context: https://github.com/easystats/parameters/issues/347#issuecomment-732931589

Maybe also leave out the internal numbers rscale_contin and rscale_discrete to avoid this confusion? At least until the moderator analysis is implemented.

Will you be open to having another quick release with these changes in the near future (pre-Christmas)?

danheck commented 4 years ago

Thanks for the suggestion! I will consider this. Maybe it is possible to push a new version of the package to CRAN before Christmas.

IndrajeetPatil commented 4 years ago

Cool, thanks for being open to this possibility :)

IndrajeetPatil commented 3 years ago

Do you this is a temporary error with the patched version? https://cran.r-project.org/web/checks/check_results_metaBMA.html

image

danheck commented 3 years ago

The error seems to have disappeared already (2020-12-01 23:48:32 CET) - maybe this was an issue with CRAN?

IndrajeetPatil commented 3 years ago

@danheck Do you think this can now be done some time in January? I am planning to have my next package release cycle then, and I'd like to do it after metaBMA defaults have changed.

danheck commented 3 years ago

@IndrajeetPatil I just uploaded a version 0.6.6 with new default priors to CRAN.

IndrajeetPatil commented 3 years ago

Thanks, Daniel!

If revdepcheck flags any of my packages for failing tests, just let the CRAN people know that you have already informed the maintainer and the necessary changes have been made.

danheck commented 3 years ago

Perfect, thanks!