# Getting the residual standard error

This term is used to calculate the *t*-statistics. The distance between the expected values is represented by the dots in *Figure 9.15*, and the straight line of the model is measured by the unexplained variation values. This best-case scenario for these unexplained variation values is to have a small residual standard deviation or a close distance between the expected values and the model:

The unexplained variation distances have several points. The residual standard error says how scattered these points are. The ideal scenario is that we have a small average of unexplained variation and also a small standard deviation of unexplained variation. This means that the linear model fits the expected values of horsepower and miles per gallon.

Use the following distance measures between the model and the expected values to do a statistical analysis of the confidence of the...