a. Holding educ, exper, tenure, and iq fixed, state the estimated difference in log(wage) between married African American and single African American people. Using the p-value approach, test the null hypothesis that the coefficient of African (β₆ ) is equal to zero against the alternative hypothesis that it is negative. Show your work. (5 marks)

b. Comment on the coefficient of exper. Explain in maximum 5 sentences. (That is, how do you comment on 0.009611) (5 marks)

c. List the steps required to test whether the variable exper is significant. In your answer, also include the null and alternative hypothesis, the name of the test to be applied, and the decision rule. (5 marks)

d. Assuming an initial IQ level of 100 with 10 years of schooling:

i. If the IQ score of a person increases 1 unit, calculate the change in monthly earnings holding the rest of the variables fixed. (5 marks)

ii. If IQ score of a person increases 2 units, calculate the change in monthly earnings holding the rest of the variables fixed. (5 marks)

e. Using output B, test if the impact of IQ on log(wage) is significant for someone with educ=10 at 0.05 significance level. 
 

Transcribed Image Text:

In this scenario, we model wages on several variables. In particular, we use the following linear model: Log(wage) Bo + B₁Educ + B₂Exper+ B3 Exper² + B.Tenure + BsMarried +B.African + B,Married African + PgIQ + B,Educ * IQ + u The following output was obtained using. EViews: Output A Dependent Variable: LOG(WAGE) Method: Least Squares Sample: 1 460 Included observations: 460 Variable C EDUC EXPER EXPER^2 TENURE MARRIED AFRICAN MARRIED* AFRICAN IQ EDUC*IQ R-squared Adjusted R-squared SE of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Coefficient Std. Error t-Statistic 0.0000 0.7840 0.6243 5.712872 0.802806 7.116133 0.016210 0.059114 0.274222 0.009611 0.019612 0.490065 -0.000134 0.000817 -0.163757 0.012919 0.003417 3.781177 0.0002 0.0003 0.8700 0.206385 0.056208 3.671785 -0.011107, 0.173244 -0.064111 0.9489 0.4508 0.142229 0.188456 0.754705 0.000987 0.007428 0.132819 0.8944 0.000291 0.000545 0.533914 0.5937 0.190888 0.174706 0.360193 58.38240 Schwarz criterion -177.9429 Hannan-Quinn criter. 11.79616 Durbin-Watson stat 0.000000 Mean dependent var S.D. dependent var Akaike info criterion Prob. 6.835596 0.396488 0.817143 0.906953 0.852508 1.830893 In this scenario, we model wages on several variables. In particular, we use the following linear model: Log(wage) Bo + B₁Educ + B₂Exper+ B3 Exper² + B.Tenure + BsMarried +B.African + B,Married African + PgIQ + B,Educ * IQ + u The following output was obtained using. EViews: Output A Dependent Variable: LOG(WAGE) Method: Least Squares Sample: 1 460 Included observations: 460 Variable C EDUC EXPER EXPER^2 TENURE MARRIED AFRICAN MARRIED* AFRICAN IQ EDUC*IQ R-squared Adjusted R-squared SE of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Coefficient Std. Error t-Statistic 0.0000 0.7840 0.6243 5.712872 0.802806 7.116133 0.016210 0.059114 0.274222 0.009611 0.019612 0.490065 -0.000134 0.000817 -0.163757 0.012919 0.003417 3.781177 0.0002 0.0003 0.8700 0.206385 0.056208 3.671785 -0.011107, 0.173244 -0.064111 0.9489 0.4508 0.142229 0.188456 0.754705 0.000987 0.007428 0.132819 0.8944 0.000291 0.000545 0.533914 0.5937 0.190888 0.174706 0.360193 58.38240 Schwarz criterion -177.9429 Hannan-Quinn criter. 11.79616 Durbin-Watson stat 0.000000 Mean dependent var S.D. dependent var Akaike info criterion Prob. 6.835596 0.396488 0.817143 0.906953 0.852508 1.830893