Data Science Programming All-in-One For Dummies

Chapter 3

Predicting Outcomes Using Bayes

IN THIS CHAPTER

Working with Bayes’ Theorem in a basic way

Performing tasks using networked Bayes

Using Bayes for linear regression

Using Bayes for logistic regression

Sometimes, a method of doing something seems less complicated than it actually is, which is how things went with Bayes’ Theorem. The theorem receives its name from the Reverend Thomas Bayes, who discovered it in the 1740s and promptly discarded it. Later, Pierre Simon Laplace worked on it to give it its modern form, and then he discarded it as well. According to Sharon McGrayne's book, The Theory That Would Not Die, many people used Bayes’ Theorem for all sorts of purposes, from mounting a defense for Captain Dreyfus (see https://www.business-standard.com/article/markets/the-bayesian-curse-112011300083_1.html for details) to breaking the German Enigma code (see https://www.investsmart.com.au/investment-news/the-theory-that-cracked-the-enigma-code/138342 for details), but they generally used it in secret. Not until the twenty-first century would anyone actually admit to using ...