Linear regression implemented four different ways

My Take

Linear regression is one of the most versatile and fundamental tools for statistical modeling. It forms the basis of many different techniques, both for modeling and inference. As with so many things, there are many ways to achieve the same goal. This piece does a great job of showing a few of those ways, and documenting that they usually come up with the same answer.

Their Take

In this post we will explore the foundation of linear regression and implement four different methods of training a regression model on linear data: Simple linear regression, Ordinary least squares regression, Gradient descent, and MCMC parameter estimation with Metropolis-Hastings

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.