The short answer is that I don't know. I have interdisciplinary projects and there are also engineers in the group. Personally, I'm not an expert on these models. With data, I work on regressions and statistical analyses. I also work on machine learning, artificial intelligence, and so on.
People are always saying that the model used is a bit outdated and that we should use a different one, for example. In my opinion, that won't solve the problem. Once again, it's an empirical question. Would a new model be useful or not?
