contrasts about sccomp HOT 4 CLOSED

Hello Ricardo,

could you please install the last github version, and if the problem persists specify

. Print out the input dataset immediately before the bug occurs
. Paste the code immediately leading to the bug
. Print out of the output, if any
. Print out of the complete error/warning message, if any
. sessionInfo()

from sccomp.

• Thanks for the reply!
• After installation with GitHub (sccomp version 1.3.4), sccomp_glm recognizes "contrasts".
• However, a new issue appears. When I run the scomp_glm, the following comes out:

sccomp says: outlier identification first pass - step 1/3 [ETA: ~20s]
Exception: Variable beta_random_intercept missing (in 'model_glm_multi_beta_binomial_generate_date' at line 45)
sccomp says: outlier identification second pass - step 2/3 [ETA: ~60s]
Exception: Variable beta_random_intercept missing (in 'model_glm_multi_beta_binomial_generate_date' at line 45)
sccomp says: outlier-free model fitting - step 3/3 [ETA: ~20s]
sccomp says: the composition design matrix has columns: (Intercept), GroupASD, GroupCOVID19, GroupEpilepsy, GroupHealthy, GroupLBD, GroupMS
sccomp says: the variability design matrix has columns: (Intercept)

• I tried to reinstall the package in its complex version (devtools::install_github("stemangiola/sccomp@cmdstanr")), and the following error appears during the installation:

installing source package ‘sccomp’ ...
using staged installation
SYNTAX ERROR, MESSAGE(S) FROM PARSER:
error in 'model_glm_multi_beta_binomial' at line 47, column 26

45:                         ) {
46: 
47: return beta_binomial_lupmf(
                             ^
48:     slice_y |

from sccomp.

I see. that's a problem with windows. I will solve it tomorrow and keep you posted.

from sccomp.

Check of #73 are almost complete. Hopefully, this release will solve your problems.

from sccomp.