Looking into the MatchIt articles made me realize that using Bootstrap with BCa is a better practice for assessing uncertainty estimation (since: "For nonlinear models (e.g., logistic regression), the delta method is only an approximation subject to error"). Using Estimating Effects After Matching article for clustered bootstrap, I got satisfactory results. Still, I would like to utilize the full functionality of the marginaleffects package, for instance - using the "by =" to get stratified results, etc. The marginaleffects documentation shows how can marginaleffects::avg_comparisons can be used in combination with the boot package using marginaleffect::inferences. But, when I try to use bootstrap after running marginaleffects::avg_comparisons with clusters G-comp according to MatchIt article, I get an error.