05. Screencast + Text: How Does MLR Work?

The text noted at the end of the video is provided below.

How Do We Find the "Right" Coefficients in Multiple Linear Regression

In the simple linear regression section, you saw how we were interested in minimizing the squared distance between each actual data point and the predicted value from our model.

But in multiple linear regression, we are actually looking at points that live in not just a two dimensional space.

For a full derivation of how this works, this article provides a breakdown of the steps.

The takeaway for us is that we can find the optimal \beta estimates by calculating (X'X)^{-1}X'y.

In the following video, you will use statsmodels to obtain the coefficients similar to how we did it in the last concept, but you will also solve for the coefficients using the equation above to show the results are not magic.