Also I could bring the call of third party software (ergm R package) to a lower level : calling theyr C++ functions to get the change statistics instead of R. Not sure if I need only the change stats. Maybe I use also the pmle, that should be easy to substitute with a full julia logistic regression
Also I could bring the call of third party software (ergm R package) to a lower level : calling theyr C++ functions to get the change statistics instead of R. Not sure if I need only the change stats. Maybe I use also the pmle, that should be easy to substitute with a full julia logistic regression