In Chapter 17, Exercise 46 we found a model for national Health Expenditures from an economic variable, Internet Users/100 people, and Primary Completion Rate. A look at leverage values and Cook’s distances identifies several countries as possible high influence points. Here are the values for those countries:
a) From the data, find the distribution of each variable and explain why each country was identified as a possible high influence point.
b) Find the regression model as in Chapter 17, Exercise 46 without these points and discuss briefly the difference in the two models. Should you report the model with or without these points?