- QUESTION 1 (Total 9 Marks)
Use the data in the worksheet entitled “ATTEND” in "Assignment 2.xlsx" for this question. The
data contains the following information collected for 680 university students in the United
States:
stndfnl = the standardized final exam score1
atndrte = the percentage of lectures attended
fresh = 1 if in 1st
second = 1 if in 2nd year of university; and 0 otherwise
priGPA = prior cumulative GPA (grade point average)
ACT= State high school graduation achievement test score2
a) To determine the effects of attending lecture on final exam performance, first estimate a
model relating the standardized final exam score (stndfnl) to the percentage of lectures
attended (atndrte). Include the binary variables fresh and second as explanatory variables.
Interpret the estimated coefficient on atndrte and discuss its significance:
(3 marks)
b) As proxy variables for student ability, add to the regression priGPA and ACT. Now what
is the effect of atndrte? Discuss why and how the effects differs from that in a).
(3 marks)
c) To test for a nonlinear effect of atndrte, add its squared term to the regression equation in
b). What do you conclude?
(3 marks)
1
student obtained a final exam score 1 standard deviation above his/her class average, and a standardized final exam
score of -3 means the student scored 3 standard deviations below the class average in the final exam.
2
Achievement test scores are often used in an educational system to determine what level of instruction for which a
student is prepared. High achievement scores usually indicate a mastery of grade-level material, and the readiness
for advanced instruction. Low achievement scores can indicate the need for remediation or repeating a course grade.
year of university; and 0 otherwise
, a student with a standardized final exam score of 1 means the
QUESTION 2 (Total 11 Marks)
Use the data in the worksheet entitled “NLS80” in “Assignment 2.xlsx” for this question. The
intention is to investigate returns to education. The data contains the following variables
collected for 935 working men in the United States:
WAGE = weekly earnings ($)
AGE = age in years
EDUC = years of education
EXPER = years of work experience
TENURE = years with current employer
MARRIED= 1 if married; and 0 otherwise
SOUTH=1 if living in the southern region; and 0 otherwise
URBAN=1 if living in an SMSA (Standard Metropolitan Statistical Area); and 0 otherwise
BLACK = 1 if black; and 0 otherwise
IQ = IQ test scores
a) Use the data in “NLS80” to estimate the following log(wage) equation for men:
(1 mark)
b) Interpret
determining wage.
(2 marks)
c) Re-estimate the regression model with the addition of the variable IQ - a proxy variable
for ability. How is the re-estimated model different from a) and why?
(3 marks)
and discuss both the economic and statistical significance of education in
d) Add the interaction variable to the regression model estimated in c). Use
your results to compare the estimated return to education between non-blacks and blacks
and interpret the statistical significance of the results.
(3 marks)
e) What other potential explanatory variables might be missing from the model. Justify your
answer with reference to existing studies on determinants of wage levels (make sure you
reference their findings and include discussions as to why these additional factors are
relevant).
(2 marks)
QUESTION 3 (Total 5 Marks)
Use the data in the worksheet entitled “NZ” in "Assignment 2.xlsx" for this question, it provides
New Zealand annual Gross National Expenditure (E) and Gross National Income (I) for the
period between 1987 and 2013.
Choosing the appropriate lag length, run a distributed lag model to evaluate the impacts of
Income ($ millions) on Expenditure ($ millions), state the initial econometric model you will
estimate and your final estimated model. Interpret the estimated coefficients found.