mboeck11
commented
2 years ago Hi,
It seems that BGVAR doesn't find any draws that fulfill the trimming argument. Have you changed the eigen argument? Do you find draws when you compute the model with 1000 instead of 5000 draws, not changing any of the other specifications? If you have changed the value of eigen, to which value?
I should built in a specific error message for this case, thanks for pointing this out.
Best, Max
Hi, I am trying to execute the BGVAR for more than 50 countries and at some point I keep getting a problem when stacking the estimated country models. The error I get is below. The interesting point is that when I set the number of draws to 1000, it does work but setting it to 5000 prompts the error. I thought that this could be due to the RAM (storage), but when I run the example model given in the vignette, it seems to work. What am I missing? Any help would really be appreciated. By the way, I am using MacOs with 4 cores + Intel Processor + 8GB RAM. " Estimation of country models starts... Estimation done and took 4 mins 41 seconds. Start stacking: 0% 10 20 30 40 50 60 70 80 90 100% [----|----|----|----|----|----|----|----|----|----| **|
Stacking finished. Computation of BGVAR yields 0 (0%) draws (active trimming).Error in l$post : $ operator is invalid for atomic vectors In addition: Warning message: In parallel::mclapply(X, FUN, ..., mc.cores = cores) : all scheduled cores encountered errors in user code "