dev

[DominiqueMakowski]

Introducing new function to convert a Bayesian model into a frequentist one. Currently only support lm/glm and lmer/glmer. Could be improved.

Why? It's part of an idea for easystats/effectsize#8. At first I thought about making it an internal in parameters but then I thought, that this convenience function could be useful for people to run quick comparisons.

[codecov-io]

Codecov Report

Merging #243 into master will decrease coverage by 0.37%.
The diff coverage is 0%.

@@            Coverage Diff            @@
##           master    #243      +/-   ##
=========================================
- Coverage   57.67%   57.3%   -0.38%     
=========================================
  Files          56      57       +1     
  Lines        2429    2445      +16     
=========================================
  Hits         1401    1401              
- Misses       1028    1044      +16

Impacted Files	Coverage Δ
R/convert_bayesian_to_frequentist.R	0% <0%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d46fe0c...0cd8a04. Read the comment docs.

[strengejacke]

Looks good! Of course these super freaky brms models will not work, neither will some rstanarm models (gamm?) for now, but the basic implementation is nice!

[DominiqueMakowski]

Can merge then (and submit?)

[strengejacke]

I just submitted insight 0.6.0, maybe you want to wait until it's accepted on CRAN, or do you want to submit bayestestR independtly from insight?

[DominiqueMakowski]

Can wait so we can update requirements

[strengejacke]

last submission wasn't even a month ago, so at some point CRAN will tell you that you should submit no more often than every 1-2 months anyway ;-)

[DominiqueMakowski]

this is a major round of update tho, after easystats is fully on CRAN the pace will (might) slow down