For the Crime2 data file at the text website, fit the linear regression model with y = violent crime rate and x = percentage living in metropolitan areas, for all 51 observations.

(a) Plot the studentized residuals. Are there any clear outliers?

(b) Identify any observations with noticeable leverage.

(c) Based on

(a) and (b), do any observations seem to be particularly influential? Explain.

(d) Study the DFFIT values. Which, if any, observations have a strong influence on the fitted values?

(e) Study the DFBETA values. For each term, which if any observations have a strong influence on the parameter estimate?

(f) Remove the observation that seems most influential, and refit the model. Is the prediction equation substantively different in any way?