Description
This course will introduce students to the fundamentals of econometrics and
and stress the practical application of these fundamentals to estimation of
economic models on real data.
The course will begin with a short primer on matrix algebra and a series of
tutorials on SAS, a statistical package widely used in the "real" world,
and Shazam, also widely used.
The emphasis on theory and practice will be roughly 2 to 1 in the course.
The practice - estimation of models - will be performed on SFU computer lab
machines and the successful student will be one that devotes much effort
toward gaining fluency in statistical programming and analysis.
Texts
D. Gujarati's Basic Econometrics and P. Kennedy's
A Guide to Econometrics.
I will arrage for a variety of other texts to be on reserve in the library.
Course Outline
(available as a text file)
Grades
The final grade for the course will be based on weekly assignments (15%),
a quiz in the second week of classes (10%),
a midterm in week 9 (25%), and a final examination (50%).
Sample Midterms
These come from econ435, a similar course, and should provide a guide.
Sample Finals
These come from econ435, a similar course, and should provide a guide.
Assignments
Assignment Lab
Accounts will be set up on the assignment labs (MAC and PC).
The MAC lab in WMX accepts your UNIX id or your assignment lab id but
requires the prefix "ECON_835_D1_" for the assignment lab id.
Data
- #1 real estate sales (available as a text file)
- #2 Qualitative data from transportation example (available as a text file)
Sample SAS Programs
- #1 shows how to do a Monte Carlo experiment in proc iml
(available as a text file)
- #2 shows how to use the cards input statement
(available as a text file)
- #3 shows how to read in data from the Mac hard drive
(available as a text file)
- #4 shows how to merge data sets
(available as a text file)
- #5 reads in real estate sales data with a cards statement
and does a regression (available as a text file)
- #6 This does GLS by reweighting observations
with a variable assumed to be proportional to a heteroskedastic effect
and with a Feasible Aitken procdeure. LM tests are also performed.
(available as a text file)
- #7 This tests for AR(1) and ARCH(1) errors with LM tests and performs ML
estimation of an AR(1) model (available as a text file)
- #8 This deseasonalizes data (available as a
text file)
- #9 This takes annual data and converts it to monthly data
(available as a text file)
- #10 This takes monthly data and converts it to annual data
(available as a text file)
SHAZAM
Links to Other Interesting Web Sites:
Mark Kamstra, Assistant Professor | kamstra@.sf
u.ca