Exercise 4: Diagnostics

Note

Diagnostics

1. This problem involves the Boston data set. We now try to predict per capita crime rate using the median value of owner-occupied homes.

   1. Fit a simple linear regression model. Check if there are any potential leverage points and outliers.

   2. Check the normality, constant variance and linearity.

   3. Transform \(y\) and/or \(x\), and fit a new linear regression. Make sure it satisfies linear regression assumptions.