Let's see how to interpret the coefficients of the model, such as the TV ad coefficient (beta1):
- A unit increase in the input/feature (TV ad) spending is associated with a 0.047537 unit increase in Sales (response). In other words, an additional $100 spent on TV ads is associated with an increase in sales of 4.7537 widgets.
The goal of building a learned model from the TV ad data is to predict the sales for unseen data. So, let's see how we can use the learned model in order to predict the value of sales (which we don't know) based on a given value of a TV ad.