Around 20:45 in the lecture, it says that it’s a problem for model fitting if n > p. But It should be the other way around.