The 311 cases that make up this data set are types of cars sold in the 2016 model year in the United States. The variables include the weights (in thousands of pounds) and urban driving mileage (in miles per gallon). For this analysis, the response is 100 divided by the miles per gallon, and the explanatory variable is the weight of the car.

(a) Generate the scatterplot of the number of gallons per 100 miles on the weight of the car (given in thousands of pounds). Fit the regression of the number of gallons per 100 miles on the weight of the car. Do you see a problem with the SRM for these data?

(b) How many residuals lie outside the 95% prediction bands? According to the SRM, how many of these should lie above and how many should lie below the estimated regression line?

(c) Does the location of these large (in absolute size) residuals anticipate that these data do not meet the conditions of the SRM? Do the residuals meet the conditions of the SRM?