I came across this thread today.
It has a very good discussion both in the comments and in the answers, yet it was closed as off-topic because it is about numerical optimization. This is very puzzling to me. We have quite a number of threads about optimization, and a whole optimization tag dedicated to it.
Optimization clearly plays a large part in classical statistics (e.g. logistic regression) and modern statistics (e.g. neural networks).
Would anyone involved with closing this question care to comment on why it was closed?