A reporter calls to interview you for an article she's writing on wages in the United States. Assuming MLR.1 through MLR.6 hold, answer this reporter's questions using the following regression of the hourly wage in dollars (wage) on years of education (educ), number of dependents (numdep). and years of experience (exper): Source 1 SS df NS
Model 1 1644.27077 3 $48.090258 Number of obs = 526
Residual 1 15516.14352 522 10.5673247 FC 3, 522) - 51.87
Prob > F -0.0000
R-squared - 0.2296
Adj R-squared - 0.2252
Root MSE - 3.2507
Total 1 7160.41429 525 13.6388844
wage 1 Coef Std. Er t P(t) (95% Conf. Interval)
reduc 1 .6677621 .0553717 12.06 0.000 .5589834 .7765409
numdep 1 .2021022 .1161096 OUTPUT REDACTED> -.0259973 .4302018
exper 1 .2725892 .0110497 6.57 0.000 .0508819 .0942965
-cons -3.939016 .8274316 -4.76 2.000- -5.564521 -2.313511
1. Using these regression results, calculate a 99% confidence interval for the returns to education. What is the lower confidence limit (LCL) of your confidence interval?
2. The reporter asks whether having more dependents affects wages. To answer her question, using a 1% size of test, you plan to test whether number of dependents has an impact on wages. Calculate the relevant test statistic.
3. Given the test statistic found in Question 2, what is the smallest significance level at which you can reject the null hypothesis?
4. How would your answer change if, instead of implementing the test exactly as described in Question 2, you decided to test whether number of dependents has a positive impact on wages?
A. The test-statistic calculated for Question 2 would change.
B. The rejection region calculated for Question 3 would change.
C. The smallest significance level at which you can reject" calculated for Question 4 would change.