Economics 320
Econometrics
Spring
2002
Mike Robinson
Office: 128 Skinner
538-2215
Office Hours: MW
9-11 and by appointment.
(I am usually
around so feel free to stop by.)
Email:
mirobins@mtholyoke.edu
Text: Pindyck and Rubinfeld, Econometrics Models and Economic
Forecasts, Irvin McGraw Hill.
Objectives: This course is an extension of previous work in
statistics with emphasis on estimation and statistical inference. Students will be exposed to econometric
theory and applications of econometric techniques. The classical regression model will be examined along with its
problems. SPSS will be the primary
programming language used. The course
goal is for each student to competently employ regression techniques as
research tools and to justify and defend the techniques used.
Lab: Each week a computer lab will be held to aid
the students in the use of the computer.
Problem sets to be done in lab will be handed out each Wednesday and due
the following Wednesday. Absolutely no
late problem sets will be accepted.
They must be in by class on Wednesday.
Course
Requirements: Lab Reports (Weekly) 10%
Mid-term
(March 12) 30%
Second
Mid-Semester Exam 30%
Final
Paper 30%
The
above percentages are suggested and will be adjusted to your benefit if
necessary.
Attendance
(though not counted per se as part of the grade) will be very important!
Research Paper: You should select a topic related to
some economic issue or problem you have been exposed to in another course and
one that interests you. One idea would
be to select a published piece of empirical research and redo the analysis with
new data or a slightly different model.
Often authors may be willing to provide you with the data from their
study if you write and explain your interest.
An advantage of this approach is the ease with which you can compare
your results with those of the original study.
We also have a great deal of data available in the department and the
Web is an excellent data source. Some example
papers can be found at the following URL
http://www.mtholyoke.edu/~mirobins/320papers.html
Consult
me if you need help with a topic.
A one-page report on the topic with
bibliography should be handed in before Spring Break (Please turn in 2 copies -
one for me and one to be returned).
First drafts are due by April 26 and final papers are due by the end of
the semester. Students will be required
to present their results in conference format during the final few days of
class. Some extra class sessions may be
necessary to provide adequate time for the presentations. Group projects are
appropriate and encouraged. You should
early on form a group of 2 or 3 for your project.
Course outline
and Readings Listed below are subjects and corresponding text
readings for the semester.
1. Least Squares and Introduction Jan.
28-Feb. 6
Chapters 1-3
2. The Multiple Regression Model Feb.
11-20
Chapter 4-5
3.
Heteroskedasticity and Serial Correlation Feb.
25 - March 6
Chapter 6
Midsemester
Exam March
8-10
4. Model Specification March
11-13
Chapter 7
5. Qualitative Dependent Variables March
25-27
Chapter 11
6. Panel Data April
1 - 3
Chapter 9
7. Simultaneous Systems April 8-10
Chapter 12
Second
Exam April
12-14
8.
Forecasting and Simulation April
15
Chapters 8, 13, 14
9.
Maximum likelihood estimation April
17
Chapter 10
8. Advanced Topics - Time Series April
22-24
Chapters 15-19
9. Student Presentations April 26 -
May 6
Final
Paper due by the end of finals
Friday
4th Hours
Fridays
11-11:50 (and 10-10:50) have been set aside as a 4th hour for extra
discussions, computer instruction, and help with problem sets and papers. The following is the initial tentative
schedule of 4th hours. We will be
meeting in Carr 101 for the computer lab assistance.
Paper Topics and Data Sources
Choosing
a topic.
Develop a hypothesis. From other classes. Talking to faculty. From the Newspaper. Check out Econlit for what others have done.
Data
sources:
The
economics department and myself maintain a number of data sources. While you may not want to choose a paper
topic simply by the availability of the data, knowing the available data may
help you know which projects are feasible and which are not feasible.
National Longitudinal
Survey of Youth. 12,000 in the sample
who were 14-21 in 1979 followed through the 1990s. Lots of great data for most any labor market study. Also lots of sociological kinds of questions
regarding drug and alcohol use, religion, time use.
Topics: Effect of working on the time allocation of women towards
childcare.
Effect of union status on earnings.
PADI: Data on wages, employment, exports and
imports for Latin America by sector.
Baccalaureate
and Beyond. Data on 1993 college
graduates in 1997.
World
Tables. All the world tables data. Annual data from many countries around the
world.
Impact of various factors on
economic growth.
The
college has access to the ICPSR, which is a very large number of datasets. See the library’s web page at
http://www.mtholyoke.edu/lits/library/ref/ICPSR.html
In addition a large amount of data
is available from the Web. For
information on this data....see internet resources at
http://netec.wustl.edu/~adnetec/WebEc/.
Most Macro data, Current Population
Survey (includes many interesting surveys), 1990 Census (lots of data good for
studying a specific occupation).