jhart96 / bram Goto Github PK

It's quite rare that we might want to z-score only some of the predictors, but auto-scaling them all is tedious. Add option to auto-scale all predictors.

ATE (causal inference literature). Rather than just contrasts in a 2-category variable, use the new_data option in the posterior to get the ATE (e.g. between yes and no). Contrasts would take the estimated distribution for the actual data for yes and also for no and then take the difference. ATE takes the difference in distributions between the actual data with all of that covariate set to yes, and the data with all covariates set to no.

Including the -o trick and give user options for cores and threads per chain