A linear regression model is fitted to a dataset (xi , yi) for i = 1, 2, . . . , n. Denote by yi the fitted value and by 2 the residual mean squares (i.e., MSRES).

(1) Define the raw residual ei and standardized residual di using yi , yi and 2 .

(2) How do you use di to identify potential outliers? Only stating how, no need to explain.

(3) When plotting ei against the fitted value yi , describe the expected shape of the plot if there are no violations of the model assumptions and no outliers..