It's currently only possible to compute the standardized mean difference via the balance calculators per_feature_loss() method. This is a bit too low level for something so commonly computed. This could be perhaps be implemented as a method to MatchingData, or as a standalone routine.
It's currently only possible to compute the standardized mean difference via the balance calculators per_feature_loss() method. This is a bit too low level for something so commonly computed. This could be perhaps be implemented as a method to MatchingData, or as a standalone routine.