MID-TERM TEST 2021

Financial Econometrics Duration: 60 minutes

Exam Code: 03


Problem 1 (2 points):
We collect data from a survey on wages, education and experience and estimate the model:
wage =  0 + 1educ +  2 exp er + u (1)
where: wage= monthly salary
educ= years of education
exper= years of work experience

The results are presented in Table 1.

Table 1: Regression results for eq. (1)

VARIABLES Model 1

educ 0.644
(0.0538)
exper 0.0701
(0.0110)
Constant -3.391
(0.767)

Observations 526

ESS 1612
RSS 5548
Standard errors in parentheses

(i) Test the hypothesis that β1 <0.5 and compute 99% confidence interval of β1.
(ii) Test the significance of the model.
F0.05;1;524 = 3.86 F0.05;2;523 = 3.013
(iii) What does the intercept mean?

Problem 2 (2.5 points):

The equation of interest is:
𝑠𝑙𝑒𝑒𝑝 = 𝛽0 + 𝛽1 𝑡𝑜𝑡𝑤𝑟𝑘 + 𝛽2 𝑒𝑑𝑢𝑐 + 𝛽3 𝑎𝑔𝑒 + 𝛽4 𝑎𝑔𝑒𝑠𝑞 + 𝛽5 𝑦𝑛𝑔𝑘𝑖𝑑 + 𝑢 (2)

where sleep= minutes sleep at night per week

totwrk= minutes worked per week
educ= number of years schooling
age= age in years
agesq= age2
yngkid =1 if age of children<3, 0 otherwise
1
(i) Table 2 provides the results of this equation for the whole sample and those for the
subsamples of men and women separately. Compute the Chow test for equality of
the parameters in the sleep equation for men and women.
Table 2: Regression results for eq. (2)

Model 2.1 Model 2.1 Model 2.1

VARIABLES male=0 male=1 Whole sample

totwrk -0.140*** -0.182*** -0.146***

(0.0277) (0.0245) (0.0169)
educ -10.21 -13.05* -11.14*
(9.589) (7.414) (5.890)
age -30.36 7.157 -8.124
(18.53) (14.32) (11.37)
agesq 0.368 -0.0448 0.126
(0.223) (0.168) (0.135)
yngkid -118.3 60.38 17.15
(93.19) (59.02) (50.01)
Constant 4,239*** 3,648*** 3,825***
(384.9) (310.0) (240.3)

Observations 306 400 706

R-squared 0.098 0.156 0.115
F-statistics 6.495 14.59 18.14
RSS 57,288,576 63,763,979 123,267,451
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
male=1 for men, male=0 for women
F0.05 ,5,694 = 2.23 , F0.05 ,11,694 = 1.80 , F0.05 ,6,699 = 2.11, F0.05 ,6,694 = 2.11

(ii) Now, allow for a different intercept for males and females and determine whether
the interaction terms involving male are jointly significant. State and test the
hypothesis, using the results presented in Table 3.

Table 3: Regression results for eq. (2) with variable male

VARIABLES Model 2.2 Model 2.3

male 87.75** -590.5

(34.67) (488.8)
totwrk -0.163*** -0.140***
(0.0182) (0.0264)
male*totwrk -0.0422
(0.0367)
educ -11.71** -10.21
2
 (5.872) (9.164)
male*educ -2.847
(11.97)
age -8.697 -30.36*
(11.33) (17.71)
male*age 37.51
(23.12)
agesq 0.128 0.368*
(0.135) (0.213)
male*agesq -0.413
(0.276)
yngkid -0.0228 -118.3
(50.28) (89.06)
male*yngkid 178.7*
(108.1)
Constant 3,841*** 4,239***
(239.4) (367.9)

Observations 706 706

R-squared 0.123 0.131
F-statistics 16.30 9.479
RSS 122,147,777 121,052,555
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1

F0.05 ,5,694 = 2.23 , F0.05 ,11,694 = 1.80 , F0.05 ,6,699 = 2.11, F0.05 ,6,694 = 2.11

(iii) Given the results from parts (i) and (ii), what would be your final model?

Problem 3 (2.5 points):

Suppose we collect data from a survey on wages, education, experience, tenure, gender and marital
status. We estimate the model:
lwage =  0 + 1educ +  2 exp er +  3tenure +  4 married +  5 female + u (3)
where
wage= monthly salary
lwage= ln(wage)
educ= years of education
exper= years of work experience
tenure= years with current employer
female=1 if female, 0 otherwise
married=1 if married, 0 otherwise

The results are presented in Table 3.

Table 3: Regression results for eq. (3)

3
 VARIABLES Model 3

educ 0.083
(0.00697)
exper 0.00313
(0.00168)
tenure 0.0169
(0.00296)
married 0.126
(0.0400)
female -0.286
(0.0373)
Constant 0.490
(0.101)

Observations 526
R-squared 0.404
F-statistics 70.38
Prob>F 0.000

Standard errors in parentheses

(i) Discuss the results in Table 3.

(ii) Extend the original model to allow the return to education to depend on marital status
and explain the test whether the return to education does depend on marital status.

Problem 4: (1 point)
Consider the standard wage equation:

lwage = 0 + 1educ + 2 exp er + 3tenure + u (4)

where
wage= monthly salary
lwage= ln(wage)
educ= years of education
exper= years of work experience
tenure= years with current employer
State the null hypothesis that another year of general workforce experience has the same effect on
wage as another year of tenure with the current employer. Test the hypothesis using the followings
results. Justify your answer.
Table 4: Regression results

VARIABLES Model 4.1 Model 4.2

educ 0.0749*** 0.0749***

(0.00651) (0.00651)
4
 exper 0.0153*** 0.00195
(0.00337) (0.00474)
tenure 0.0134***
(0.00259)
sum 0.0134***
(0.00259)
Constant 5.497*** 5.497***
(0.111) (0.111)

Observations 935 935

R-squared 0.155 0.155
F-statistics 56.97 56.97
ESS 25.70 25.70
RSS 140 140
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
sum=exper+tenure

Problem 5: (2 points)
In an analysis of the returns to two-and four year colleges, we estimate the model:
lwage = 0 + 1 jc + 2univ + 3 exp er + u (5)

where jc = number of years attending a two-year college.

univ = number of years at a four-year college
exper = months in the workforce
The hypothesis of interest is whether one year at a junior college is worth one year at university.
The alternative of interest is one-sided: a year at a junior college is worth less than a year at a
university. Explain the process to test the hypothesis.