Rj editor: long vectors not supported yet

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)
Post Reply
MaximeB
Posts: 5
Joined: Fri Feb 25, 2022 1:53 pm

Rj editor: long vectors not supported yet

Post by MaximeB »

Hello Jamovi Team,

First of all, thank you for making Jamovi freely available. It has been collaborative projects a lot easier than other statistical packages (eg, SPSS).

When doing linear mixed-effect models (LMM), I am used to throwing robust LMM as sensitivity analyses to safeguard myself against potential bias from outlying observations. I normally do this in R with the robustlmm package (function rlmer). However, I wanted to import my script on Jamovi to make it available to my colleagues unfamiliar with R. Running the rlmer function works as intended. However, when I try to compute confidence intervals, I get the following errors:

Code: Select all

long vectors not supported yet: array.c:2192
Given that the code runs correctly on R, I was wondering if the issue came from Jamovi. If this is a known issue, is there a workaround?

I am unable to attach a screenshot to my post, but here's the simple script I have tried to use:

Code: Select all

# Make sure Jamovi is using system R (and not Jamovi R)       ->
library(lme4)
library(robustlmm)

# Custom functions ----
# Function given by Ben Bolker on Cross Validated to compute CIs from rlmer
# https://stats.stackexchange.com/questions/233800/how-can-i-get-confidence-intervals-for-fixed-effects-using-the-rlmer-function-r
confint.rlmerMod <- function(object,parm,level=0.95) {
  beta <- fixef(object)
  if (missing(parm)) parm <- names(beta)
  se <- sqrt(diag(vcov(object)))
  z <- qnorm((1+level)/2)
  ctab <- cbind(beta-z*se,beta+z*se)
  colnames(ctab) <- stats:::format.perc(c((1-level)/2,(1+level)/2),
                                        digits=3)
  return(ctab[parm,])
}



# Write the formula you want to use in the function below
# Please, make sure the variables are listed above (in the right box)
rlmer <- rlmer(
    data = data,
    formula = score ~ group * condition + (1|id)
    )

summary(rlmer)
confint(rlmer)
Thank you in advance for your help!
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Rj editor: long vectors not supported yet

Post by jonathon »

could you attach a .omv file (with the analysis)? that would make it easier.

cheers

jonathon
MaximeB
Posts: 5
Joined: Fri Feb 25, 2022 1:53 pm

Re: Rj editor: long vectors not supported yet

Post by MaximeB »

Hello jonathon,

Thanks for your reply. Here's a watered-down version of the .omv file. Don't mind the model being tested, it is just for testing purposes.

PS: I am getting a "there is no package called rappdirs" error message from my personal computer (that would not go away with installing jvmRedWrite). Hopefully, you get the original error message (long vectors not supported yet).

Regards,
Attachments
rpe-ems_LMM-robust - Copy.omv
(4.09 KiB) Downloaded 295 times
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: Rj editor: long vectors not supported yet

Post by jonathon »

hi,

sorry for the delay.

i notice that only the call to `confint(rlmer)` produces this error. you might need to reach out to the author of robustlmm to ask why this occurs.

cheers

jonathon
Post Reply