Consider the following estimated equation that explains salaries of CEOs using data on 42 firms: log(salary)=(.47)4.37+(.049)(.272log(sales)+(.0065).0168roe(.00135).0017ros, n=42,R2=0.469, where salary is CEO salary (in thousands $ ), sales is firm sales (in millions \$), roe is return on equity (in percentages), and ros is return on firm's stock (in percentages). Standard errors in parentheses. 1 (a) Interpret the coefficient on log (sales). (b) Find the 95% confidence interval for the coefficient on log (sales). Can you reject the hypothesis that H0:log(sales)=1 against H1:log(sales)=1 at the 5% significance level? (c) Interpret the coefficient on ros. Does ros have a practically (or economically) large effect on salary? Is ros statistically significant at the 10% level? (d) Test the hypothesis that roe has no effect on salary against the alternative that roe has a positive effect. Do the test at the 5% and 1% levels. Instructions: lways start by writing down the null and alternative hypotheses when you answer such hypothesis testing questions. Consider the following estimated equation that explains salaries of CEOs using data on 42 firms: log(salary)=(.47)4.37+(.049)(.272log(sales)+(.0065).0168roe(.00135).0017ros, n=42,R2=0.469, where salary is CEO salary (in thousands $ ), sales is firm sales (in millions \$), roe is return on equity (in percentages), and ros is return on firm's stock (in percentages). Standard errors in parentheses. 1 (a) Interpret the coefficient on log (sales). (b) Find the 95% confidence interval for the coefficient on log (sales). Can you reject the hypothesis that H0:log(sales)=1 against H1:log(sales)=1 at the 5% significance level? (c) Interpret the coefficient on ros. Does ros have a practically (or economically) large effect on salary? Is ros statistically significant at the 10% level? (d) Test the hypothesis that roe has no effect on salary against the alternative that roe has a positive effect. Do the test at the 5% and 1% levels. Instructions: lways start by writing down the null and alternative hypotheses when you answer such hypothesis testing questions