I have a dataset on "effects of marijuana on mental health" and it has 9 lac observations. Now I assume this to be a large file since it is not opening in stata. Is there any way I can reduce the size of the dataset without cleaning it beforehand?