From the course: Machine Learning and AI Foundations: Prediction, Causation, and Statistical Inference

Unlock the full course today

Join today to access over 24,400 courses taught by industry experts.

Explain vs. predict

Explain vs. predict

- [Instructor] Galit Shmueli does a remarkable job expanding on the kinds of issues Breiman raises in the "Two Cultures," but I think she's gone beyond that. Not just expanding it by addressing more of the differences, but also modernizing the conversation. She's written a paper, but she has also given keynotes all around the world discussing this important topic. Her basic premise is to go out into the world, frankly, almost like an anthropologist would, examining papers that have been submitted to various journals and conferences, and try to better understand and describe how analytics actually occurs in the wild. She describes three groups. One group is focused on good explanatory power and assumes that if I have a model that explains well, it will predict well too. They aren't statisticians exactly as she describes it. They are applying statistical ideas, but in areas like the social sciences, business, and industry.…

Contents