Refer to the Chief Executive (Sept. 1999) study of CEOs, Exercise 10.8 (p. 574). Recall that a CEO's pay y was modeled as a function of company performance x₁ and company sales x₂ using a first-order model. Refer to the EXCEL printout on p. 575 and answer the following

a. Find and interpret the value of the multiple coefficient of determination

b. Give the null and alternative hypotheses for testing whether the overall model is statistically useful for predicting a CEO's pay.

c. Give the value of the test statistic and corresponding p-value for the test of part h

d. Conduct the test of part b using α .05. What is your conclusion?

Exercise 10.8

Refer to the Chief Executive (Sept. 1999) study of chief executive officers (CEOs) from a variety of industries, exercise 9.35 (p. 500). Recall that a CEO's pay (y) was modeled as a function of company performance (x₁), where performance was measured as a three-year annualized total return to shareholders assuming dividends are reinvested. For this exercise, consider a second independent variable, company sales (x₂). The data for all three variables are listed in the table below.

a. Construct a scattergram of total pay versus company sales. Does your scattergram suggest that company sales will help explain the variation in CEO pay? Explain.

d. Locate a 95% confidence interval for ß₁ on the printout and interpret it in the words of the problem

Exercise 9.35

Some critics of big business argue that CEOs are overpaid and that their compensation is not related to the performance of their companies. To test this theory, Chief Executive (Sept. 1999) collected data on executive's total pay and company's performance for each in a sample of 17 CEOs selected from a variety of industries. The data are listed in the table below. (Note: Company performance is the three-year annualized total return to shareholders for the years 1996 to 1998. assuming dividends are reinvested.)

a. Construct a scattergram for these data. Does it appear that CEO pay is related to company performance? Explain.

b. Use the method of least squares to model the relationship between CEO pay (y) and company performance (x).

c. Is CEO compensation related to company performance? Conduct the appropriate hypothesis test using a = .05.

d. Interpret the estimate of ß₁ in the context of the problem. 

e. Construct a 90% confidence interval for ß₁, and interpret your result in the context of the problem.

f. How might the results of part e change if a sample of CEOs from the same industry were used'?