Member-only story
Detect Change Points with Bayesian Inference and PyMC3
Update Your Beliefs Using Observed Data
Motivation
Imagine you use Google Analytics to track the views of your website. After looking at the views, you suspect there might be a sudden change in the number of views after a certain date.
If there is indeed a date when the number of views changes suddenly, how do you find that date?
Wouldn’t it be nice if you can leverage observed data and some beliefs to guess the change point with high certainty?
That is when Bayesian inference comes in handy. In this article, we will learn what Bayesian inference is and how to use PyMC3 to perform Bayesian analysis.
What is Bayesian Inference?
Bayesian inference is a technique in which Bayes’ theorem is used to specify how one should update one’s beliefs upon observing data.