Hello,
I am new to brms and bayesian analysis. I am trying to run a mixed model with two (fixed effects) predictors e.g. A ~ 1+ B + C + (1|ID) - it seems to run and produces a summary, conditional effects plot, and hypothesis estimates, however I am concerned that it does not create a plot when I run plot.brmsfit, while for a single predictor e.g. A~ 1+ B+ (1|ID) it does.
I am not sure if this is expected behaviour or if it indicates I am doing something wrong.
I have tried to create a minimal reproducible example using the Iris dataset; this gives a warning that ‘parts of the model do not converge’ which is not the case in my dataset, but does reproduce the issue with plot.brmsfit - hopefully this is enough to demonstrate my query.
#setup
if(!require(brms)){
install.packages("brms",repos = "http://cran.us.r-project.org")
}
library(brms)
iris$ID <- c(1:150)
#single predictor
m <- brm(Sepal.Length ~ 1 + Petal.Length + (1|ID), data=iris, iter=10000)
#plot.brmsfit produces expected output
plot(m)
#as do other functions
plot(conditional_effects(m))
summary(m)
hypothesis(m, "Petal.Length>0")
#additional predictor
m <- brm(Sepal.Length ~ 1 + Petal.Length + Sepal.Width + (1|ID), data=iris, iter=10000)
#plot.brmsfit now creates empty plot
plot(m)
#other functions still behave as expected
plot(conditional_effects(m))
summary(m)
hypothesis(m, "Petal.Length>0")
- Operating System: macOS Mojave 10.14.6
- brms Version: 2.12.0
Thanks!