(1)Suppressor variables.Agresti & Finlay p. 316, problem 10.20 (7.5 points)

When we analyze data for the census tracts in the greater Los Angeles area, we find no significant correlation between median tax bill and median lot size. Yet a considerable positive correlation occurs when we control for the percentage of the tract used for business. Explain how the percentage of the tract used for businesses could be a suppressor variable if it is positively correlated with median tax bill and negatively correlated with median lot size.

(2)Statistical interactions.Agresti & Finlay p. 316, problem 10.22 (7.5 points)

For lower-level managerial employees of a fast-food chain, the prediction equation relating Y = annual income (thousands of dollars) to = number of years experience on the job equals = 14.2 + 1.1for males and = 14.2 + 0.4for females. Explain how these equations show evidence of statistical interaction.

(3) Multiple regression. Agresti & Finlay, problem 11.2 (parts a, b, c, and d) (2.5 points for each answer, a-d)

For recent data in Jacksonville, Florida, on y = selling price of homes (in dollars), X1 = size of home (in square feet), and X2 = lot size (in square feet), the prediction equation is = -10,536 + 53.8+ 2.84

(a)A particular home of 1240 square feet on a lot of 18,000 square feet sold for $145,000.Find the predicted selling price and the residual, and interpret.

(b)For fixed lot size, how much is the house selling price predicted to increase for each square-foot increase in home size? Why?

(c)According to this prediction equation, for fixed home size, how much would lot size need to increase to have the same impact as a one-squared-foot increase in home size?

(d)Suppose house selling prices are changed from dollars to thousands of dollars.Explain why the prediction equation changes to = -10,536 + 0.0538+ 0.00284.