I am doing a competing risks regression using STATA 13.1 MP2 on a dataset with approx 4 million observations and 1 independent variable, which is an interaction term. It takes days. Anything to speed it up (apart from more processors + better STATA)?

Similar questions and discussions