Minimizing the squared error (MSE, or Mean Squared Error) as we did in class is not the only way to estimate 0 and 1. In this question, we'll explore the Mean Relative Error (MRE). Surprisingly, this has nothing to do with the errors that my meanest relatives make. Instead, the MRE is defined as 1 =1 ( )2 , which results in different estimators for 0 and 1 than the MSE. This is often used for financial applications when the bank is less worried about whether their model fits for those with a lot in thei