You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Although easy to generate a data frame with rvars from {emmeans} (similar to what's possible with marginaleffects::posterior_draws()), it would be nice to have this option built in - it will fit nicely with the other *_rvars() alternatives to the *_draws() functions.
library(dplyr)
library(brms)
library(emmeans)
library(tidybayes)
set.seed(5)
n<-10ABC= tibble(
condition= rep(c("A","B","C","D","E"), n),
response= rnorm(n*5, c(0,1,2,1,-1), 0.5)
)
m<- brm(response~condition, data=ABC,
chains=1, iter=500)
eml<- emmeans(m, pairwise~condition)
gather_emmeans_draws(eml)
#> # A tibble: 3,750 × 7#> # Groups: condition, contrast, .grid [15]#> .grid condition .chain .iteration .draw .value contrast#> <chr> <chr> <int> <int> <int> <dbl> <chr> #> 1 emmeans A NA NA 1 -0.271 NA #> 2 emmeans A NA NA 2 -0.00172 NA #> 3 emmeans A NA NA 3 -0.167 NA #> 4 emmeans A NA NA 4 0.0142 NA #> 5 emmeans A NA NA 5 0.0934 NA #> 6 emmeans A NA NA 6 -0.0276 NA #> 7 emmeans A NA NA 7 -0.0110 NA #> 8 emmeans A NA NA 8 0.249 NA #> 9 emmeans A NA NA 9 0.162 NA #> 10 emmeans A NA NA 10 0.165 NA #> # ℹ 3,740 more rows#> # ℹ Use `print(n = ...)` to see more rowsgather_emmeans_rvars<-function(object, ...) {
tidybayes::nest_rvars(tidybayes::gather_emmeans_draws(object))
}
gather_emmeans_rvars(em_condition)
#> # A tibble: 15 × 4#> # Groups: condition, contrast, .grid [15]#> condition contrast .grid .value#> <chr> <chr> <chr> <rvar[1d]>#> 1 A NA emmeans 0.1914 ± 0.18#> 2 B NA emmeans 1.0134 ± 0.18#> 3 C NA emmeans 1.8733 ± 0.18#> 4 D NA emmeans 1.0220 ± 0.18#> 5 E NA emmeans -0.9509 ± 0.17#> 6 NA A - B contrasts -0.8220 ± 0.27#> 7 NA A - C contrasts -1.6819 ± 0.26#> 8 NA A - D contrasts -0.8306 ± 0.24#> 9 NA A - E contrasts 1.1423 ± 0.25#> 10 NA B - C contrasts -0.8600 ± 0.25#> 11 NA B - D contrasts -0.0086 ± 0.26#> 12 NA B - E contrasts 1.9643 ± 0.24#> 13 NA C - D contrasts 0.8514 ± 0.25#> 14 NA C - E contrasts 2.8242 ± 0.25#> 15 NA D - E contrasts 1.9729 ± 0.26
The text was updated successfully, but these errors were encountered:
Although easy to generate a data frame with
rvar
s from{emmeans}
(similar to what's possible withmarginaleffects::posterior_draws()
), it would be nice to have this option built in - it will fit nicely with the other*_rvars()
alternatives to the*_draws()
functions.The text was updated successfully, but these errors were encountered: