26.Does the revised model offer a statistically significant improvement over the initial simple regression? a)Yes, because R2has gotten larger. b)No, because the increase in R2is too small to be significant. c)Yes, because the RMSE has gotten smaller. d)No, because the t-statistic for Yeardecreased. e)Yes, because the p-value for Year2is less than 0.05.

Statistics 621, Final Exam -8 of 14- Fall, 2009 Questions 27-38 A banking analyst was interested in predicting the yield on a ten-year bond to be issued by a company in South America. To do this, the analyst collected data on 84 previous issues from different companies in the region. After looking at many models, the analyst settled on one that includes a country effect, a measure of the financial leverage of the company (the ratio of the total liabilities of the company to its net worth), together with an interaction between the country and financial leverage. The country effect is a categorical variable with 5 levels (Argentina, Brazil, Caribbean, Chile, and Mexico. (Note: Be careful here, as “leverage” is a variable name that is being used in the financial sense rather than in the statistical sense!) Output from the multiple regression follows. The yield (the response in the multiple regression) is reported on a unit basis so that, for example, 5% is coded as 0.05. Summary of Fit RSquare 0.65461 Root Mean Square Error 0.01527 Mean of Response 0.08529 Observations 84 Analysis of Variance Source DF Sum of Squares Mean Square F Ratio Model 9 0.03272175 0.003636 15.5834 Error 74 0.01726488 0.000233 Prob > F C. Total 83 0.04998664 <.0001* Effect Tests Source Nparm Sum of Squares F Ratio Prob > F Financial Leverage 1 0.00454001 19.4592 <.0001* Country 4 0.00149662 1.6037 0.1824 Country*Financial Leverage 4 0.00068691 0.7361 0.5703 ArgentinaBrazilCaribbeanChileMexicoCountry

Statistics 621, Final Exam -9 of 14- Fall, 2009 Indicator Function Parameterization Term Estimate Std Error t Ratio Prob>|t| Intercept 0.06528 0.00498 13.12 <.0001* Financial Leverage 0.00851 0.00184 4.63 <.0001* Country[Argentina] 0.01646 0.00745 2.21 0.0302* Country[Brazil] 0.00273 0.00671 0.41 0.6858 Country[Caribbean] 0.00705 0.00950 0.74 0.4603 Country[Chile] -0.00639 0.01359 -0.47 0.6399 Country[Argentina]*Financial Leverage 0.00046 0.00242 0.19 0.8489 Country[Brazil]*Financial Leverage -0.00263 0.00259 -1.01 0.3139 Country[Caribbean]*Financial Leverage -0.00059 0.00296 -0.20 0.8418 Country[Chile]*Financial Leverage -0.00740 0.00653 -1.13 0.2613 Least Squares Means Table Level Least Sq Mean Std Error Mean Argentina 0.1012 0.0037 0.1057 Brazil 0.0808 0.0029 0.0792 Caribbean 0.0895 0.0050 0.0945 Chile 0.0613 0.0055 0.0609 Mexico 0.0838 0.0035 0.0816 Financial Leverage, Leverage Plot Residuals 27.The intercept in the fitted model estimates that the yield on bonds issued by a company a)Averages 6.528% if the Financial Leverageratio equals 0. b)Increases by 6.528% if the Financial Leverageratio increases above 0. c)Averages 6.528% in South America if the Financial Leverageratio equals 0. d)Increases by 6.528% if the company relocates from Mexico to Argentina. e)Averages 6.528% for Mexico if the Financial Leverageratio equals 0.

#### You've reached the end of your free preview.

Want to read all 14 pages?

- Fall '08
- STAFF
- Regression Analysis