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
First of all, many thanks for the wonderful package!
I have a question and/or feature idea. I estimated models with brms, drew posterior samples and used "describe_posterior", which really worked fine. But is there a possibiliy to set different upper and lower ROPE bounds (or also different values for pd) for different variables of the posterior?
For example: I calculated a simple t-test like model with two groups and different standard deviations for each group. Within the posterior draws, I also calculated the effect sizes, so I have 5 variables in my posterior object. If I use
post |>
describe_posterior(centrality = "all",
ci_method = "hdi",
test = c("p_direction", "rope"),
rope_range = c(80, 100))
for example, the upper and lower bounds of 100 and 80 will be used for all variables, which is not useful, for this application.
It would be very helpful, if I could enter a list instead of a vector, so that each variable could get it's own ROPE bounds. Maybe something like this:
post |>
describe_posterior(centrality = "all",
ci_method = "hdi",
test = c("p_direction", "rope"),
rope_range = list(
c(80, 100), # ROPE for group 1
c(100, 120), # ROPE for group 2
c(5, 20), # ROPE for SD 1
c(10, 30), # ROPE for SD 2
c(-0.1, 0.1) # ROPE for standardized effect size
)
As far as I found, a list is only possible for multivariate analyses, where I can submit different ROPE bounds for each dependent variable, but these would be the same for every variable/predictor within each dependent variable. Is this correct?
Please correct me, if this option is already available and I missed it.
Many thanks for help!
Rainer
The text was updated successfully, but these errors were encountered:
Hello!
First of all, many thanks for the wonderful package!
I have a question and/or feature idea. I estimated models with brms, drew posterior samples and used "describe_posterior", which really worked fine. But is there a possibiliy to set different upper and lower ROPE bounds (or also different values for pd) for different variables of the posterior?
For example: I calculated a simple t-test like model with two groups and different standard deviations for each group. Within the posterior draws, I also calculated the effect sizes, so I have 5 variables in my posterior object. If I use
for example, the upper and lower bounds of 100 and 80 will be used for all variables, which is not useful, for this application.
It would be very helpful, if I could enter a list instead of a vector, so that each variable could get it's own ROPE bounds. Maybe something like this:
As far as I found, a list is only possible for multivariate analyses, where I can submit different ROPE bounds for each dependent variable, but these would be the same for every variable/predictor within each dependent variable. Is this correct?
Please correct me, if this option is already available and I missed it.
Many thanks for help!
Rainer
The text was updated successfully, but these errors were encountered: