The SLR model used in Assignment 2, to examine the relationship between years of education and the annual salary of a particular bank's employees, was linear in both the parameters and the variables: salaryi|educ = Bo + Bieduci + U. (3) An alternative model is: log(salaryi) educ = Yo + Vieduci + Ei (4) (b) Estimate model (3), summarizing your sample results in the usual way. Provide your EViews/R output. What is your prediction of the estimate average change in annual salary between two employees with all factors the same, except that one has 6 more years of education? Ensure that you provide sufficient details to justify your answer. Do you think this estimated change in average annual salary, ceteris paribus, is practically relevant? Explain. 3 marks Dependent Variable: SALARY Method: Least Squares Date: 10/14/20 Time: Sample: 1 474 Included observations: 474 Variable Coefficient Std. Error t-Statistic Prob. EDUC -18331.18 3909.907 2821.912 204.5470 -6.496014 19.11495 0.0000 0.0000 R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.436338 Mean dependent var 0.435144 S.D. dependent var 12833.54 Akaike info criterion 7.77E+10 Schwarz criterion -5155.528 Hannan-Quinn criter 365.3814 Durbin-Watson stat 0.000000 34419.57 17075.66 21.76172 21.77928 21.76863 1.365464 (c) Detail the properties you can claim regarding this samples estimate of y ? Explain. 1 mark (d) Put in plain words how you might use the standard error reported for the partial effect of educ from your sample regression function. The SLR model used in Assignment 2, to examine the relationship between years of education and the annual salary of a particular bank's employees, was linear in both the parameters and the variables: salaryi|educ = Bo + Bieduci + U. (3) An alternative model is: log(salaryi) educ = Yo + Vieduci + Ei (4) (b) Estimate model (3), summarizing your sample results in the usual way. Provide your EViews/R output. What is your prediction of the estimate average change in annual salary between two employees with all factors the same, except that one has 6 more years of education? Ensure that you provide sufficient details to justify your answer. Do you think this estimated change in average annual salary, ceteris paribus, is practically relevant? Explain. 3 marks Dependent Variable: SALARY Method: Least Squares Date: 10/14/20 Time: Sample: 1 474 Included observations: 474 Variable Coefficient Std. Error t-Statistic Prob. EDUC -18331.18 3909.907 2821.912 204.5470 -6.496014 19.11495 0.0000 0.0000 R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) 0.436338 Mean dependent var 0.435144 S.D. dependent var 12833.54 Akaike info criterion 7.77E+10 Schwarz criterion -5155.528 Hannan-Quinn criter 365.3814 Durbin-Watson stat 0.000000 34419.57 17075.66 21.76172 21.77928 21.76863 1.365464 (c) Detail the properties you can claim regarding this samples estimate of y ? Explain. 1 mark (d) Put in plain words how you might use the standard error reported for the partial effect of educ from your sample regression function
