A Bayesian, Frequentist, and Machine Learner practitioner meet at a bar…
Authors: Dror Berel
Read the closeread article: A Bayesian, Frequentist, and Machine Learner meet at a bar… – Dror Berel. Statistical Consultant. There is R package for that...
Reproducible repo: Blog/close_dialog at main · drorberel/Blog · GitHub
Abstract
In his influential 2001 paper, "Statistical Modeling: The Two Cultures," Leo Breiman explored two dominant paradigms in data analysis: Data Modeling, often aligned with frequentist approaches, and Machine Learning (ML). This post addresses key questions that arise from Breiman's work: Why did he choose the term "Data Modeling" rather than "Frequentist"? Why did he omit Bayesian statistics as a potential third culture? What distinguishes these approaches in philosophy and practice? To provide clarity, the post includes a fictional dialogue between three practitioners, illustrating the fundamental differences between these methodologies.