Closed
Description
It would be nice if these graphs could be interactive to show the bayesian posterior probability.
The user should get the area to the right of point x (probability of effect great than x)
Is that possible?
library(tidyr)
library(tidybayes)
library(ggplot2)
set.seed(1234)
pp2<-tribble(
~y, ~x,
"a1", exp(rnorm(8000, 0.16, 0.04)),
"a2", exp(rnorm(8000, 0.24, 0.1)),
"b1", exp(rnorm(8000, 0.24, 0.1)),
"b2", exp(rnorm(8000, 0.24, 0.1)),
"c1", exp(rnorm(8000, 0.16, 0.04)),
"c2", exp(rnorm(8000, 0.19, 0.09)),
"c3", exp(rnorm(8000, 0.18, 0.07)),
"d1", exp(rnorm(8000, 0.16, 0.04)),
"d2", exp(rnorm(8000, 0.24, 0.08)),
"d3", exp(rnorm(8000, 0.17, 0.09)),
"d4", exp(rnorm(8000, 0.2, 0.1)),
) %>%
unnest(x) %>%
ggplot(aes(x, y ) ) +
geom_halfeyeh()+
geom_vline(xintercept = c( 1), linetype = "dashed") +
scale_fill_manual(values = c("blue", "gray80"), limits=c(0.1,0.7) ) + theme_tidybayes()
ggplotly(pp2) # doest not work
Metadata
Metadata
Assignees
Labels
No labels