22.

(a) Draw a scatter diagram of cigarettes smoked versus death rate from leukemia.

(b) Estimate the regression coefficients.

(c) Test the hypothesis that there is no regression of the death rate from leukemia on the number of cigarettes used. That is, test that β = 0.

(d) Determine a 90 percent prediction interval for the leukemia death rate in a state whose citizens smoke an average of 2,500 cigarettes.