14.31 - Econometrics,
Fall 1997
Professor Dora L. Costa, costa@mit.edu
TA: Steven Bergantino, sberg@mit.edu
Office hours will be by appointment (made via e-mail)
- Course Description and Prerequisites.
This course will introduce econometric analysis of linear models, both
theory and application. The only formal prerequisite is 14.30 or the equivalent,
but familiarity with matrix algebra will be helpful.
- Texts. The required texts are Econometric
Models and Economic Forecasts by Pindyck and Rubinfeld and The Practice
of Econometrics, Classic and Contemporary by Berndt. The latter will
also be a good place to look for ideas for the empirical paper.
- Requirements The course grade will be
based on two exams (30 percent each), 6-7 problem sets (20 percent), and
an empirical paper (20 percent). One sheet of notes will be allowed for
the exams. Bring a calculator. We will provide any necessary tables. You
may colloborate on the problem sets, but must write up and hand in your
own solutions. You should feel free to consult any sources, including your
classmates, for help on your empirical paper, but each student will, of
course, be required to hand in his or her own original paper. Futhermore,
the deadline for the paper will be strictly enforced. You will lose at
least 1/4 of a grade if you hand it in later. The later you hand it in
the more your grade will suffer. Note that the deadline is close to the
last exam. Therefore try to get the paper in a bit earlier. To give you
even more of an incentive, I will give you extra credit (1/4 of a grade)
if you hand it in Thursday the 4th in class. Regular attendace at the recitation
is strongly recommended. The TA will discuss problem sets, clarify lecture
material, and provide computer guidance. Problem sets will be posted on
the class web page, http://web.mit.edu/14.31/www/
- Important Dates
- October 21 or 23, Exam 1 in class
- November 11, Holiday (Veterans' Day)
- November 19, Drop Date
- November 27, Holiday (Thanksgiving)
- December 5, paper due
- December 9, Exam 2 in class
- Course Outline Chapters in parentheses
are from Pindyck and Rubinfeld (PR) and Berndt (B)
- Review of Probability and Statistical Inference
(PR: ch 2): Random variables, expectation and variance, point and interval
estimation, hypothesis testing
- Simple Linear Regression (PR: ch 1, 3): Least
squares estimation, statistical properties of estimates, goodness of fit
- Multiple Regression (PR: ch 4, 5) (B: ch 3, 4,
5): Estimation of regression coefficients, tests of linear restrictions,
dummy variables, prediction
- Heteroskedasticity and Autocorrelation (PR: ch
6)(B: ch 6): Generalized least squares estimation
- Specification Error (PR: ch 7)(B: ch 8, 10):
Omitted variables, nonlinearities, measurement error, simultaneous equations
- Advanced Topics (B: ch 11): Panel data, discrete
choice models, time series