This weekend is about Linear models, linear models are the simplest parametric methods because many problems, even intrinsically non linear ones, can be easily solved with these

Models, we can try to identify the decision surface that separates the two classes. linear regression, is one of these linear models and a very simple approach for supervised learning.

Linear Regression is a useful tool for predicting a quantitative(interval, ratio) response(output). It is not a time to waste, it’s time to learn and understand linear regression or even compare it with modern and advanced statistical learning algorithms, linear regression is still useful learning algorithm, and it’s used as a base for a newer and powerful algorithms, also many fancy techniques can be seen as a generalization of linear regression. Hence it is very important to have a good understanding of linear regression before deep diving to complex learning methods.

In this weekend, we review the following:

• Some of the key ideas underlying linear regression.
• The least square approach that most commonly used to fit this model.