Thanks for the read!

A few things I wanted to mention that cause a lot of confusion in linear regression. The first is, technically, the linear relationship is between the predicted value (output) and the model parameters, which are slightly different from the inputs (the inputs afterall, can be of non-linear).

The second comment, and perhaps a future candidate for this "whats the difference" blog series, which I think is a great idea, would be comparing a residual and an error. These are often confused to be the same thing, but there is a subtle difference. Residuals are the deviation between an observed value and the predicted value of that observation, whereas errors are the deviation between an observed value and the true value of that observation, which we often will never know.