1. Slide 10- 1 Copyright © 2010 Pearson Education, Inc. Active Learning Lecture Slides For use with Classroom Response Systems Business Statistics First Edition by Sharpe, De Veaux, Velleman Chapter 17: Understanding Residuals

2. Slide 17- 2 Copyright © 2010 Pearson Education, Inc. Examining and understanding residuals for an estimated regression model enables us to I. Check if separate subgroups exist in the data. II. Check that the linearity condition is met. III. Check to see if the data consist of summary values. A. I only B. II only C. I and II D. I, II and III

3. Slide 17- 3 Copyright © 2010 Pearson Education, Inc. Examining and understanding residuals for an estimated regression model enables us to I. Check if separate subgroups exist in the data. II. Check that the linearity condition is met. III. Check to see if the data consist of summary values. A. I only B. II only C. I and II D. I, II and III

4. Slide 17- 4 Copyright © 2010 Pearson Education, Inc. Using a linear regression model to predict y for a value of x that is not included in the range of x values used in fitting the model is called A. estimation. B. extrapolation. C. forecasting. D. prediction.

5. Slide 17- 5 Copyright © 2010 Pearson Education, Inc. Using a linear regression model to predict y for a value of x that is not included in the range of x values used in fitting the model is called A. estimation. B. extrapolation. C. forecasting. D. prediction.

6. Slide 17- 6 Copyright © 2010 Pearson Education, Inc. In fitting a linear regression model, points with large residuals are called A. outliers. B. forecasts. C. predictions. D. extrapolations.

7. Slide 17- 7 Copyright © 2010 Pearson Education, Inc. In fitting a linear regression model, points with large residuals are called A. outliers. B. forecasts. C. predictions. D. extrapolations.

8. Slide 17- 8 Copyright © 2010 Pearson Education, Inc. Which of the following statements is true about influential points? A. A high leverage point is always an influential point. B. An influential point always has a large residual. C. A residual plot is the best way to identify an influential point. D. An influential point always affects the slope of the estimated regression equation.

9. Slide 17- 9 Copyright © 2010 Pearson Education, Inc. Which of the following statements is true about influential points? A. A high leverage point is always an influential point. B. An influential point always has a large residual. C. A residual plot is the best way to identify an influential point. D. An influential point always affects the slope of the estimated regression equation.

10. Slide 17- 10 Copyright © 2010 Pearson Education, Inc. A large multinational corporation interested in the relationship between salary and years of experience among its employees fit a regression using data at the division rather than the individual level. They used mean salaries and mean years of experience for a sample of divisions. This would result in A. a higher correlation than if data were collected at the employee level. B. a lower correlation than if the data were collected at the employee level. C. the same correlation as if the data were collected at the employee level. D. a zero correlation.

11. Slide 17- 11 Copyright © 2010 Pearson Education, Inc. A large multinational corporation interested in the relationship between salary and years of experience among its employees fit a regression using data at the division rather than the individual level. They used mean salaries and mean years of experience for a sample of divisions. This would result in A. a higher correlation than if data were collected at the employee level. B. a lower correlation than if the data were collected at the employee level. C. the same correlation as if the data were collected at the employee level. D. a zero correlation.

12. Slide 17- 12 Copyright © 2010 Pearson Education, Inc. The Durbin Watson statistic I. is used to check the assumption of independent errors in regression. II. is always conclusive regarding the autocorrelation in residuals. III. should be used when the data are time series. A. I only B. I and II C. I and III D. I, II and III

13. Slide 17- 13 Copyright © 2010 Pearson Education, Inc. The Durbin Watson statistic I. is used to check the assumption of independent errors in regression. II. is always conclusive regarding the autocorrelation in residuals. III. should be used when the data are time series. A. I only B. I and II C. I and III D. I, II and III

14. Slide 17- 14 Copyright © 2010 Pearson Education, Inc. A simple regression model was fit using quarterly GNP to predict quarterly S&P500 Index values over a 10 year period. The Durbin Watson statistic was calculated as 1.02. The lower critical value from the table is d L = 1.25. We can conclude that A. There is evidence of positive autocorrelation it the residuals. B. There is no evidence of autocorrelation in the residuals. C. There is evidence of negative autocorrelation in the residuals. D. The test is inconclusive.

15. Slide 17- 15 Copyright © 2010 Pearson Education, Inc. A simple regression model was fit using quarterly GNP to predict quarterly S&P500 Index values over a 10 year period. The Durbin Watson statistic was calculated as 1.02. The lower critical value from the table is d L = 1.25. We can conclude that A. There is evidence of positive autocorrelation it the residuals. B. There is no evidence of autocorrelation in the residuals. C. There is evidence of negative autocorrelation in the residuals. D. The test is inconclusive.

16. Slide 17- 16 Copyright © 2010 Pearson Education, Inc. Which of the following is not a goal of re- expressing data for use with linear regression? A. Make the form of the scatterplot more linear. B. Make the distribution of a variable more symmetric. C. Make the variance more constant. D. Make extrapolations more accurate.

17. Slide 17- 17 Copyright © 2010 Pearson Education, Inc. Which of the following is not a goal of re- expressing data for use with linear regression? A. Make the form of the scatterplot more linear. B. Make the distribution of a variable more symmetric. C. Make the variance more constant. D. Make extrapolations more accurate.