Regression Statistics Multiple R R Square Adjusted R Square Standard Error .490 Observations 1000 ANOVA df SS MS Significance F Regression 4975177517 829196252 52.2 4.3E-56 Residual 15739811503 15850767 Total t Stat P-value Lower 95% Upper 95% -6873.8 1610.0 Coefficients Standard Error -5628.1 634.8 -8.87 3.45E-18 -4382.4 Intercept CEOANN 328.2 1.15E-11 2898.3 2254.2 16.3 3.9 4.18 3.11E-05 EMPL 8.43 1.18E-16 680.8 1093.8 105.2 49.6 887.3 МKTVAL EPSIN -3.54 0.00042 -272.7 -78.2 -175.5 0.14 -123.1 876.7

*Based on the regression results, answer the following questions* A sample of data is collected (from 1999 and 2000) concerning the compensation of the executives (compensation is measured in 1000’s of $’s) of a number of public companies along with other firm-specific data. The dependent variable is total compensation, CEOANN is a dummy variable =1 for an individual who is a CEO and =0 for individuals who are not CEOs, EMPL is total employees, MKTVAL is the natural logarithm of the market value of the firm, EPSIN is earnings per share, YEAR is a dummy variable = 1 for the year 2000 and =0 for the year 1999, and ASSETS is the natural logarithm of the total assets of the company. Based on the regression results, answer the following questions h) Is there a significant amount of evidence (say at the 10% level of significance) that CEO’s make at least $2 million more than non-CEO’s? i) Suppose you’re not sure if all the variables above should be included in the regression. First of all, do you think all the variables above SHOULD be included in the regression? If your answer is NO, then how would you go above deciding which ones to eliminate?