Instructor:
Professor Colin Cameron
SSH Building 1124   752-8396
accameron@ucdavis.edu

Meeting:
Mon Wed 10.00 – 11.50 a.m. Wellman 115
3 hours lecture and one hour discussion

Office Hours:
Wed 8.30 a.m. - 10.00 a.m.
Fri  10.30 a.m. - noon

Teaching Assistant:
Zhiyuan Li: SSH 0120@ucdavis.edu
Office hours: Monday 9:00-10:00 a.m. and Tuesday 4:00-5:00 p.m.

Course Goals: (1) To be able to perform estimation and testing in basic linear cross-section regression models, (2) to understand the theory underlying ordinary least squares (OLS) using matrix algebra. This course is the foundation for any subsequent regression / econometrics class.

Pre-requisites: A solid foundation in statistics is required (the listed pre-requisite is ECN/ARE 239). Additionally I assume previous exposure to linear regression and to matrix algebra. If your preparation in these areas is thin then work hard in the first two weeks to come up to speed.

Course Outline:
 

Class 0		Basic two variable linear regression model is assumed.
Classes 1-3	3 classes	Least Squares Regression with Matrix Algebra Greene  2.1-2.2,  3.1-3.6,  Appx A.1-A.4
Class 4-5	2 classes	Finite Sample Properties of Least Squares Greene 2.3,  4.1-4.8
Classes 6-7	2 classes	Large Sample Properties of Least Squares  Greene 5.1-5.3,  Appx D.1-D.4
Class 8-9	2 classes	Tests of Linear Restrictions Greene 6.1-6.5,  Appx B.10-B.11
Class 10	1 class	Midterm Exam

 

Class 11-12	2 classes	Practical Issues Greene 4.9 (data problems), 6.6 (prediction), 7.2-7.3 (data transformation and indicator variables)
Class 13	1 class	Model Specification Error Greene 8.1-8.2
Class 14	1 class	Instrumental Variables (IV) Estimation Greene 5.4
Class 15	1 class	Maximum Likelihood (ML) Estimation Greene 17.6
Class16	1 class	Generalized Least Squares (GLS) Greene 10.1-10.3, 10.5-10.6
Class17	1 class	Heteroskedasticity Greene 11.2, 11.4-11.6
Class18	1 class	Autocorrelation Greene 12.3, 12.5, 12.7-12.8

Comparison to Previous Years

I now directly begin with regression with matrix algebra (a change from Winter 2005 and 2006).
You should try assignment 1 as soon as possible to see whether you need to work hard in the first two weeks to catch up.
I am now teaching from Greene and try to follow his sequence of topics (a change from Winter 2005 and 2006 - notably large sample theory appears much earlier in the quarter.)
I will present a much more complete presentation of GLS, heteroskedasticity and autocorrelation than in previous classes.

Required Material:

Greene, W.G. (2003), Econometric Analysis, 5th edition, Prentice-Hall.

Greene is the standard text for this course at any economics Ph.D. program. Greene is as much a reference source as it is a textbook, and the chapter sections given above include in places more material than we will cover. What you really need to know is the material I cover in lectures. 

Especially for those with a thin background in econometrics a more introductory book will be helpful.
A book I like is Johnson, J. and J. Dinardo (1997), Econometric Methods, 4th edition, McGraw-Hill, but this is out of print.
Undergraduate texts such as Stock and Watson or Wooldridge are good, but they do little matrix algebra which is a big part of this course.

Additional Material:

I have provided a review of bivariate regression on the course website.
I may also make available some other lecture notes (especially on asymptotic theory).

Computer Materials:

This course will use STATA (www.stata.com ), the leading all-purpose econometrics package for analysis of cross-section data and short panels. Often complete programs will be provided and the assignments will concentrate on interpretation of the output. Stata is available on both PC and Unix platforms for ECN and ARE students.

Course Grading:

Assignments 18%
Best 6 out of 7. Last assignment is compulsory. Each worth 3%.
Due Wednesdays Jan 10, 17, 24, Feb 5 (Mon), 23, March 7, 14.

Midterm 32%
Wednesday Feb 7   10.00 – 11.50 p.m.  Closed book exam.

Final 50%
Friday  March 16  8.00 – 10.00 a.m. Comprehensive.  Closed book exam

Assignments must be handed in on time, so solutions can be discussed in class and distributed in a timely manner.
No credit for late assignments.
Academic integrity is required. What is academic integrity? See the UCD Student Judicial Affairs website http://sja.ucdavis.edu/integ.htm.
As an exception to their rules, I permit some collaboration with other students in doing assignments, but the work handed in must be your own. Each person must create their own Stata output and write up their own answers. And you are to write on your assignment the name of the person(s) you worked with.