I cannot add save_pars(all = TRUE) to a brm() call without an error. The error I get is:
Error in formula.default(object, env = baseenv()) : invalid formula
In addition: Warning message:
Argument ‘autocor’ should be specified within the ‘formula’ argument. See ?brmsformula for help.
The model runs fine without it, but obviously I need to add it in order to compare models using add_criterion.
This issue is in part a duplicate of the issue raised by BianaS:
I have tried the solutions suggested in the thread above, but to no avail. All dependencies of arms have been updated, brms has been newly installed.
priors1 <- c(set_prior("student_t(3, 0, 2.5)", class = "Intercept"),
set_prior("student_t(3, 0, 2.5)", class = "b", coef = "Fixed"))
#runs:
fit1 <- brm(GB020 ~ Fixed, data = GB136_GB020_compl3, family = bernoulli, iter = 20000,
prior = priors1)
#doesn't run:
fit1 <- brm(GB020 ~ Fixed, data = GB136_GB020_compl3, family = bernoulli, iter = 20000,
prior = priors1, save_pars(all = TRUE))
> head(GB136_GB020_compl3)
Language_ID GB136 GB020 GB070 GB072 Glot_Family area Fixed
1 abau1245 1 0 1 1 sepi1257 N Coast New Guinea 0
2 abee1242 1 1 0 1 atla1278 African Savannah 0
4 abkh1244 1 1 0 1 abkh1242 Greater Mesopotamia 0
7 acha1250 1 0 1 1 araw1281 NE South America 0
8 ache1245 1 0 0 0 narr1279 African Savannah 1
10 achi1256 1 1 0 0 maya1287 Mesoamerica 1
- Operating System: MacOS 10.14.5
- brms Version: 2.14.4