3 In a multiple regression with 5 parameters the studentized

3)         In a multiple regression with 5 parameters, the studentized deleted residual for the last case 25 is 5.9. Is it an outlier? Why/why not?

Solution

Explanation:

Outliers can affect visual resolution of remaining data in plots (forces observations into “clusters”) – Temporary removal of outliers, and/or transformations can “spread out” clustered observations and bring in the outliers (if not removed)

• More importantly, separated points can have a strong influence on statistical models—deleting outliers from a regression model can sometimes give completely different results – Unusual cases can substantially influence the fit of the OLS model—Cases that are both outliers and high leverage exert influence on both the slopes and intercept of the model – Outliers may also indicate that our model fails to capture important characteristics of the data.