2023 Assignment

Introduction to Econometrics Assignment
Your assignment must be submitted electronically to MyLo by 4.00 pm on May 26th.
Read instructions before attempting the assignment.
Instructions
 Write or type your answers to each question on A4-sized paper.

 For submission, scan all your answers.
 Show all of the steps you take to reach your answer. No marks will be given if
all necessary steps are not shown.
 Answers must be in sentence form so that single-word or single-number answers
without explanation will be considered incomplete.
 Explain all hypothesis tests in full detail, meaning:
A. Statement of the null and alternative hypotheses.
B. Values of the test statistics and the formula used to calculate it.
C. Distribution of the test statistics.
D. Significance level of the test.
E. Statement of the rule used to reject/accept the null hypothesis.
F. Conclusion of tge test expressed in terms of the economic model.
 Attach a summary of your computer output (i.e., Eviews) at the end of your
assignment.
 Your answer must be readable without reference to the computer output, but you
should highlight those parts of your output that you explicitly use in your
answers.
Question 1
Use the data in htv_.wf1 and Eviews to answer this question. The data set includes
information on wages, education, parents education and several other variables for 1230
working men. The variables in this data set include:
educ = highest grade completed

motheduc = highest grade, mother
fatheduc = highest grade, father
abil = a measure of cognitive ability
(a) Compute the means of educ, motheduc, and fatheduc. What do you conclude?
EDUC MOTHEDUC FATHEDUC
Mean 13.0374 12.17805 12.44715
[1 mark]
(b) Draw scatter plots of (1) educ and motheduc and (2) educ and fatheduc. What do you
conclude?
Question 2
[2 marks]
(c) Write a linear regression model (i.e., population regression function) in which the
dependent variable is educ and the explanatory variables are motheduc and fatheduc.
educ = β0 + β1 * motheduc + β2 * fatheduc + ε
where:
β0 is the intercept or constant term

β1 is the coefficient for the variable motheduc, which represents the effect of mother's
education on education attainment of the individual
β2 is the coefficient for the variable fatheduc, which represents the effect of father's
education on education attainment of the individual
ε is the error term or residual, which represents the unexplained variation in the
dependent variable educ that is not accounted for by the explanatory variables
motheduc and fatheduc.
[2 marks]
(d) Estimate the regression model in (c) and report your results in equation form
along with the standard errors, number of observations and R2.
Dependent Variable: EDUC

Method: Least Squares
Date: 05/26/23 Time: 14:04
Sample: 1 1230
Included observations: 1230
Variable Coefficient Std. Error t-Statistic Prob.
MOTHEDU 0.304197145035065 0.0319266404325688 9.52800360180549 8.234432292509896

C 1 7 2 e-21
FATHEDUC 0.190285753723656 0.0222839230758268 8.53914964057990 3.945530683931066
7 8 e-17
C 6.964354532528924 0.3198205019732087 21.7758226553979 3.747659611659206
e-89
Question 3
R-squared 0.249251011617791 Mean dependent var 13.03739837398374
8
Adjusted R- 0.248027296885302 S.D. dependent var 2.354346027992775
squared 6
S.E. of 2.041603169066695 Akaike info criterion 4.267783844361775
regression
Sum 5114.312074430271 Schwarz criterion 4.280258891796815
squared
resid
Log -2621.687064282492 Hannan-Quinn criter. 4.272477370349161
likelihood
F-statistic 203.6839183187359 Durbin-Watson stat 1.748204459434113
Prob(F- 4.130785111268008
statistic) e-77
[3 marks]
(e) Interpret the estimated slope parameters in (d), while paying attention to the units of
measurement. Do these estimates comply with your expectations? Explain.
[4 marks]
(f) How well does the model fit the data? Use the R2 to explain. [2 marks]
(g) Conduct a hypothesis test to determine if the coefficient of motheduc and fatheduc are
the same. Use the 5% level of significance. What do you conclude?
[4 marks]
(h) Add the variable abil to the regression model in (c). Conduct a hypothesis test to
determine if ‘ability’ helps to explain variations in education. What do you conclude?
[3 marks]
(i) A data scientist claims that this regression analysis is sufficient to uncover the effect
of parents’ education on their children’s education. Do you agree or disagree with this
statement? Explain.
[3 marks]
[Total marks: 24]

Question 4
To estimate the relationship between hours of training per employee at the firm level and
other characteristics of firms, a student considers the following regression model:
hrsempi  1  2 granti  3 ln(salesi )  4 ln(employi )  ei
where each variable is defined as:
hrsemp = hours of training per employee at the firm level

grant = dummy variable 1, if the firm received a job training grant; 0 otherwise
nogrant = dummy variable 1, if the firm did not receive a job training grant; 0
otherwise
sales = annual sales ($)
employ = number of employees at the firm
The regression results in Eviews are provided below:
Regression results
Dependent Variable: HRSEMP
Sample: 1 320
C 11.57259 24.86771 0.465366 0.6420

GRANT 34.36007 3.724866 9.224512 0.0000
LOG(SALES) 1.516521 2.053283 0.738584 0.4607
LOG(EMPLOY) -6.655931 2.234891 -2.978190 0.0031

Adjusted R-squared 0.236991 S.D. dependent var 26.49305
S.E. of regression 23.14177 Akaike info criterion 9.133577
Sum squared resid 169231.2 Schwarz criterion 9.180681
Log likelihood -1457.372 Hannan-Quinn criter. 9.152387
F-statistic 34.02726 Durbin-Watson stat 0.868270
Prob(F-statistic) 0.000000
(a) Interpret the estimated coefficients of grant and ln(sales), while paying attention to
the units of measurement.
[4 marks]
(b) What happen if you add the dummy variable nogrant to the model? Will this
improve the regression results? Explain your reasoning.
[2 marks]
(c) Test the null hypothesis that, after controlling for ln(sales) and ln(employ), grant has
no effect on hrsemp against that the alternative that there is an effect. Carry out the
test at the 5% significance level. What do you conclude?
[3 marks]
(d) Test the null hypothesis that, after controlling for grant and ln(sales), ln(employ) has
no effect on hrsemp against the alternative that ln(employ) has a positive effect.
Carry out the test at the 1% significance level. What do you conclude?
[3 marks]
(e) The sample means of hrsemp and sales are 15.78 and 6,330,290,
respectively. Calculate the elasticity of hrsemp with respect to sales. Interpret
the result.
[2 marks]
(f) Use the regression results below to test the overall significance of the regression at
the 5% level.
[4 marks]
Regression results
Dependent Variable: HRSEMP
Sample: 1 320
C 14.96754 1.301909 11.49661 0.0000

Adjusted R-squared 0.000000 S.D. dependent var 25.71064
S.E. of regression 25.71064 Akaike info criterion 9.334248
Sum squared resid 257143.5 Schwarz criterion 9.344418
Log likelihood -1819.178 Hannan-Quinn criter. 9.338279
Durbin-Watson stat 1.384509
(g) What kind of factors are contained in the error term, ei? Are these factors likely to be
correlated with any of the explanatory variables included in the model? If so, how
does it influence the estimation results reported on the previous page?
[3 marks]
[Total marks: 21]
*********************************

2023 Assignment

Uploaded by

Copyright:

Available Formats

2023 Assignment

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2023 Assignment

Uploaded by

Copyright:

Available Formats

Introduction to Econometrics Assignment

Your assignment must be submitted electronically to MyLo by 4.00 pm on May 26th.

Read instructions before attempting the assignment.

 Write or type your answers to each question on A4-sized paper.

educ = highest grade completed

β0 is the intercept or constant term

Dependent Variable: EDUC

Variable Coefficient Std. Error t-Statistic Prob.

MOTHEDU 0.304197145035065 0.0319266404325688 9.52800360180549 8.234432292509896

[Total marks: 24]

hrsempi  1  2 granti  3 ln(salesi )  4 ln(employi )  ei

where each variable is defined as:

hrsemp = hours of training per employee at the firm level

The regression results in Eviews are provided below:

Variable Coefficient Std. Error t-Statistic Prob.

C 11.57259 24.86771 0.465366 0.6420

R-squared 0.244167 Mean dependent var 15.77676

Variable Coefficient Std. Error t-Statistic Prob.

C 14.96754 1.301909 11.49661 0.0000

R-squared 0.000000 Mean dependent var 14.96754

[Total marks: 21]

You might also like