Unformatted text preview: PADP 8130 Fertig Spring 2012 UGA Practice Problems for the Midterm 1. What is the difference between the standard deviation and the standard error? Explain using equations and words. 2. What does the Central Limit Theorem say about the sampling distribution? 3. What does the Gauss Markov Theorem state? 4. Describe 2 procedures you can do to see if the inverse of a square matrix exists? 5. True or False. In a model with dummy variables, the omitted category matters because it determines the sign of the OLS coefficient. Explain your answer. 6. Using the regression below, explain what is wrong with the following statement: “Female headed households have significantly lower family incomes because female heads have a lower level of education than male heads on average.” Source 
SS
df
MS
Number of obs =
7874
+F( 9, 7864) = 509.38
Model 
2691.3033
9
299.0337
Prob > F
= 0.0000
Residual  4616.56119 7864 .587049999
Rsquared
= 0.3683
+Adj Rsquared = 0.3676
Total  7307.86448 7873
.92821853
Root MSE
= .76619
lnfaminc 
Coef.
Std. Err.
t
P>t
[95% Conf. Interval]
+agehd 
.0588569
.0028292
20.80
0.000
.0533109
.0644028
age2  .0005824
.0000277
21.05
0.000
.0006366
.0005282
black  .2795837
.0198465
14.09
0.000
.3184881
.2406794
hisp  .0913031
.0334106
2.73
0.006
.1567966
.0258095
femalehd  .2232375
.0239106
9.34
0.000
.2701086
.1763663
marriedhd 
.557765
.0231511
24.09
0.000
.5123827
.6031472
hsdropout  .3412199
.0250994
13.59
0.000
.3904214
.2920185
somecoll 
.2085378
.0226856
9.19
0.000
.164068
.2530075
collgrad 
.4386241
.0242743
18.07
0.000
.3910401
.4862082
_cons 
9.152484
.0668183
136.98
0.000
9.021503
9.283466
 7. If I ran the following command after the regression listed above, what did I just ask Stata to test? Write out the hypothesis and interpret the results. . test
( 1)
( 2)
( 3) hsdropout somecoll collgrad
hsdropout = 0
somecoll = 0
collgrad = 0
F( 3, 7864) = 278.58
Prob > F =
0.0000 8. The estimated coefficient on years of completed education on log family income is 0.127 with a standard error of 0.004. a. What is the t statistic (assuming that the null hypothesis is that the effect of education is 0)? b. What is the 95% confidence interval on this point estimate. c. Finally, show the formula for the marginal effect of education on family income and calculate the actual value. 9. A referee writes in his/her report that he/she is concerned about whether you should combine single and married households into one regression. Assume that your dependent variable is log family income, your independent variable is education of the head, and your control variables include age and age squared. What test should you run to defend your approach? Describe the test procedure and intuition. ...
View
Full
Document
This note was uploaded on 03/28/2012 for the course PADP 8130 taught by Professor Fertig during the Spring '12 term at LSU.
 Spring '12
 Fertig

Click to edit the document details