Professional Documents
Culture Documents
Spring 19 Exam 1
Spring 19 Exam 1
Introduction to Econometrics
(ECON 3543-201)
TR @ 12:30 PM; DB 175
Professor Owen
Spring 2019
This exam consists of five (5) multiple choice questions (each worth 10 points), one essay question worth 25 points and one problem worth 25
points.
For the multiple choice questions, select the letter that best answers the question.
Be sure to explain and interpret each aspect of all models in the essay question and the problem. Remember the governing KISS principle
when you prepare your responses.
Return your work electronically to the D2L dropbox (Exam 1), or if you want to turn-in a hard copy, bring it before 5:00 pm to the Dean’s
Suite (second floor of Dillard, immediately to your left as you exit the elevator) and ask the receptionist to deliver it to my mailbox.
~1~
Multiple Choice Questions (10 points each):
1) Consider the following regression line: TestScore = 698.9 – 2.28 * STR. You are told that the t-statistic for the slope coefficient is (- 4.38).
What is the standard error of the slope coefficient? (STR = student – teacher ratio)
a. 0.52
b. 1.96
c. -1.96
d. 4.38
2
2) The regression R is defined as follows:
a. S SE
SST
RSS
b.
SST
n
(Y Y )( X
i 1
i i X)
n n
(Yi Y ) 2 (X i X )2
c. i 1 i 1
SSR
d. n 2
3) One of the following steps is not required as a step to test for the null hypothesis:
a. connecting the Yi corresponding to the lowest Xi observation with the Yi corresponding to the highest Xi observation.
b. making sure that the standard error of the regression equals the standard error of the slope estimator.
c. minimizing the sum of absolute residuals.
d. minimizing the sum of squared residuals.
1) You have obtained a sub-sample of 1744 individuals from the Current Population Survey (CPS) and are interested in the relationship between
weekly earnings and age. The regression yielded the following result:
·
Earn = 239.16 + 5.20×Age , R2 = 0.05, SER = 287.21.,
where Earn and Age are measured in dollars and years respectively. (SER = standard error of the regression, aka standard error of the estimate)
~3~
(c) Why should age matter in the determination of earnings? Do the results suggest that there is a guarantee for earnings to rise for everyone as they
become older? Is the relationship between age and earnings is linear?
We could imply that an increase in age would mean more experience and training, which leads to a higher pay. This would imply a positive
effect between age and earnings. However, the results do not suggest that there is a guarantee for earnings to increase with age since R 2 is only
0.05. The relationship between age and earnings can be seen as an inverted U-shape rather than linear. As we age, we expect to make more
money by advancing our career. However, at some point we retire and our earnings are reduced as we get older.
(d) The average age in this sample is 37.5 years. What is annual income in the sample?
Substituting the average age into the equation the weekly income would be:
·
Earn = 239.16 + 5.20×(37.5)
·
Earn = $434.16
The weekly income of someone who is 37.5 years is $434.16. Assuming 52 weeks a year, that person earns $22,576.32 annually.
Problem: (parts a, b, & c = 3 points each; part d = 16 points). This problem continues on the next page.
SUMMARYOUTPUT
Regression Statistics
0.91317312
Multiple R 2
R Square 0.83388515 CARSPEND = total expenditure on new automobiles in the United States ($, billions)
0.83292494
Adjusted R Square 9 DISPOSINC = national disposable income in the United States ($, billions)
13.0376050
Standard Error 3
Observations 175
ANOVA
df SS MS F Significance
~4~
F
147618.070 868.448131
Regression 1 147618.0707 7 1 2.43543E-69
169.979144
Residual 173 29406.39206 9
Total 174 177024.4628
Standard
Coefficients Error t Stat P-value Lower 95% Upper 95%
12.9160451
Intercept 20.1767114 1.562143144 6 3.73025E-27 17.0933981 23.2600247
0.01292997 29.4694440
DISPOSINC 9 0.000438759 2 2.43543E-69 0.012063969 0.013795989
b) Briefly comment on the goodness of fit for this model with respect to the data
The R2 implies that national disposable income explains 83.39% of the variations in total expenditure on new automobiles. This means that
16.61% of the variations are explained by other factors. Furthermore, the t-statistics for both the intercept and the independent variable are above
the critical level of 2 which implies that they are statistically significant. The P values and the significance of F is less than 0.05, and the
confidence intervals do not imply 0. These also show that the data is statistically significant.
d) The DISPOSINC for 1975 through 1979 (in billions of dollars) are given below:
1975 $430.2999878
1976 438.2999878
1977 445.7000122
1978 451.8999939
~5~
1979 460.1000061
Predict CARSPEND for each of those years.
Substituting the DISPOSINC values into the equation, the results are as follows:
1975: ^
CARSPEND=20.177+ 0.013(430.2999878) = $25.77089984 billion
^
1976: CARSPEND=20.177+ 0.013(438.2999878)=$25.87489984 billion
1977: ^
CARSPEND=20.177+ 0.013 ( 445.7000122 )=$25.97110016 billion
^
1978: CARSPEND=20.177+ 0.013(451.8999939)= $26.05169992 billion
1979: ^
CARSPEND=20.177+ 0.013(460.1000061)= $26.15830008 billion
~6~