Shanghai University Of Engineering

Sciences
Khalid sbabou 史文心

1.Use the data in GPA1 to answer this question. It is a sample of Michigan State
University undergraduates from the mid-1990s, and includes current college GPA,
colGPA, and a binary variable indicating whether the student owned a personal
computer (PC).
(i) How many students are in the sample, and how many owned their own PC? (2%)

THE NUMBER OF STUDENT IS 141

NUMBER OF STUDENT HAVING PC IS 56 SO ITS 40%

(ii) What is the average skipped (average lectures missed per week)? Is the average a
good measure of the "typical" student in this case? Explain.
THE AVERAGE IS 1.07
The average is not a ggood typical measure because it’s 34.04 %

(iii) Among students who have their own PC, what is the average colGPA? How does
this compare with the whole sample and why? (2%)
AVERAGE colGPA 3.15
And The average whole sample is 3.05
Therefore the difference is likely due to differing study habits or access to
information resources

(iv) Find the number of students graduated from Michigan high school (gradMI) in
the sample. Are students graduated from Michigan State high school “typical” in the
sample? (2%)
Find the number of students graduated from Michigan high school (gradMI) in
the sample 123
And it’s typical because it’s 87.23%

Econometrics 考试试卷 A
第1页共8页
(v) Report the average colGPA and its standard deviation.(2%)
the average colGPA and its 3.05 and he standards deviation is 0.37

2.Use the data in WAGE1 to answer this question.

(i) Find the average hourly wage and the average tenure in the sample. (2%)
the average hourly wage IS 5.90 and the average tenure in the sample IS 5.10
(ii) How many observations are in their first year job (that is, tenure= 0)?
What is the longest tenure in current job?
(iii) In the first year is 163
AND THE LARGEST IS 44

(iv) Estimate the simple regression model, lwage=β 0 + β 1 educ +u ,report

your results in the usual form. (2%)
(1) (2)
lwage lwage
educ 0.0827*** 0.0827***
(10.94) (10.94)
_cons 0.584*** 0.584***
(6.00) (6.00)
N 526 526
R2 0.186 0.186
t statistics in parentheses
*
p < 0.1, ** p < 0.05, *** p < 0.01

(v) What is the (approximate) predicted percentage increase in wage given

one more year of education?
predicted percentage increase in wage given one more year of education
8%

(vi) Does your finding in part (iii) imply that one more year of education has
a causal effect on wage? Explain (2%)
(vii) NO because the results the finding in part (iii) does not necessarily
imply a causal effect of one more year of education on wages. The
relationship estimated in part (iii) reflects correlation, not causation.

(viii) the data in WAGE2 for this exercise.

(i) Obtain the descriptive statistics for the variables wage,
educ, exper and tenure. (2%)

Econometrics 考试试卷 A
第2页共8页
(ix) Summary statistics

N Mean Std. min Median max

Dev.
wage 935 957.95 404.36 115 905 3078
educ 935 13.47 2.2 9 12 18
exper 935 11.56 4.37 1 11 23
tenure 935 7.23 5.08 0 7 22

(ii) Estimate the model lwage=β_0+β_1 educ+β_2 exper+β_3

tenure+u, and write the results in equation form. (2%)

lwage= educ0.exper0.tenure0

(iii) Interpret the intercept. Does it have a useful meaning? (2%)

lwage for an individual with 0 years of education, 0 years

of experience, and 0 years of tenure.

The intercept does not have a practical or meaningful

interpretation because individuals with 0 education, 0
experience, and 0 tenure are unlikely to exist in the
dataset.

(iv) What is the predicted lwage if educ = 13 exper=11 and

tenure= 5? What do you make of this result? Are there any
observations in the sample with these values of the explanatory

There is 630 observation

(v) If obs A has educ = 9 exper= 20 and tenure=10 and obs B

has educ = 12 exper= 16 and tenure=14, what is the predicted
difference in their hourly wage

lwage=(12-9)(16-20)(14-10)
=3

Econometrics 考试试卷 A
第3页共8页
 四、Further Analysis（3*10%）
1.Use the data in WAGE1 to answer this question.
(i)

Estimate the regression model (1)lwage=β_0+β_1 educ+β_2

exper+β_3 tenure+u and (2) lwage=β_0+β_1 educ+β_2
exper+β_3 tenure+β_4 exper^2+β_5 tenure^2+u by OLS and
report the results in the usual table form. (2%)
Ans:

∆ WAGE
=β 1+ β 3∗exper
∆ educ

(ii)For regression (1), test the null hypothesis that educ has zero
effect on lwage. (2%)

H 0 : β 3=0

(iii)For regression (2), test the null hypothesis that exper and its
quadratic form has zero effect on lwage. (2%)

T=2.09 and P=0.036 (It’s less than 5%, so we can reject no

hypothesis)

(iv) For regression (2), test H0:β_1=β_2+β_3 against a two-sided

alternative. What is the p-value of the test? (2%)

Econometrics 考试试卷 A
第4页共8页
(v) Is marginal effect of exper on lwage increasing or decreasing?
explain(2%)

 The marginal effect of exper on lwage is increasing

and its is positive. This indicates that as experience
(exper) increases, the effect of additional experience
on wages becomes larger.

The part 4

1 2.Use the data in WAGE2 for this exercise.

Consider a model in which the return to education depends upon

the amount of work experience (and vice versa): lwage=\beta_0+\
beta_1educ+\beta_2exper+\beta_3educ\astexper+u

(i) What’s the return to another year of education, holding

exper fixed? (2%)

Econometrics 考试试卷 A
第5页共8页
 (1) (2)
lwage lwage
educ 0.0920*** 0.0845***
(12.56) (11.80)
exper 0.00412** 0.0293***
(2.39) (5.54)
tenure 0.0221*** 0.0371***
(7.13) (5.13)
exper2 -0.000592***
(-5.19)
tenure2 -0.000616**
(-2.47)
_cons 0.284*** 0.202**
(2.73) (1.99)
N 526 526
R2 0.316 0.367
t statistics in parentheses
*
p < 0.1, ** p < 0.05, *** p < 0.01

(ii) State the null hypothesis that the return to education

does not depend on the level of exper. What do you think
is the appropriate alternative? (2%)

T statistic =12.56

(iii)Use the data in WAGE1 to test the null hypothesis in (ii) against
your stated alternative. (3%)

: H 0 : β 2=β 4=0(reject it's not 0)

(iv) Let \theta_1denote the return to education (in decimal form),

when exper = 12, \theta_1=\beta_1+12\beta_3. Obtain \widehat{\
theta_1} and a 95% confidence interval for\ \theta_1. (3%)

H 0 : β 12= β2 + β 2 P value is 9% SO we cant reject the

hypothesis

3.Use the data in GPA2 for this exercise.

(i) Consider the equation: colgpa=\beta_0+\beta_1hsize+\beta_2{\

rm hsize}^2+\beta_3hsperc+\beta_4sat+\beta_5female+\
beta_6athlete+u, where colgpa is cumulative college grade point
average; hsize is size of high school graduating class, in hundreds;
hsperc is academic percentile in graduating class; sat is combined
SAT score; female is a binary gender variable; and athlete is a

Econometrics 考试试卷 A
第6页共8页
binary variable, which is one for student athletes. What are your
expectations for the signs of the coefficients in this equation?(2%)

the expectation is negative

(ii) Estimate the equation in part (i) and report the results in the
usual form. What is the estimated GPA differential between
athletes and nonathletes? Is it statistically significant?(2%)

the p value is 0,936 is bigger than 0,5% so there is no

statistically significant difference in GPA between women

and women non-athlete

(iii) Drop sat from the model and reestimate the equation. Now,
what is the estimated effect of being an athlete? Discuss why the
estima te is different than that obtained in part (ii). (2%

Is NOT significant

(1) (2)
colgpa hsize
hsize -0.0569***
(-3.48)
hsizesq 0.00468** 0.131***
(2.08) (201.22)
hsperc -0.0132*** -0.00752***
(-23.07) (-14.74)
sat 0.00165***
(24.64)
female 0.155*** -0.0395**
(8.60) (-2.37)
athlete 0.169*** 0.0257
(4.00) (0.65)
_cons 1.241*** 1.539***
(15.62) (91.60)
N 4137 4137
R2 0.293 0.908
t statistics in parentheses
*
p < 0.1, ** p < 0.05, *** p < 0.01

(iv) In the model from part (i), allow the effect of being an athlete
to differ by gender and test the null hypothesis that there is no
ceteris paribus difference between women athletes and women
nonathletes.(2%)

Econometrics 考试试卷 A
第7页共8页
THE ANSWER IS 0.178 with 5%

(v) Does the effect of sat on colgpa differ by gender? Justify your
answer.(2%)

There is no effect in the sat for female it is 0.001662 and

the SAT foe non-female is 0.0016407

Econometrics 考试试卷 A
第8页共8页