Faculty Jim West, W. H. Smith Professor of Economics

ECO 5347

Graduate Econometrics

Fall 2011

Faculty Jim West, W. H. Smith Professor of Economics

Office Hankamer 345

Office Phone 710-6126

Office Hours MO, TU, WE 10-12 and by appointment

Prerequisites ECO 4347 or equivalent

Introduction

The field of econometrics can be described as a tight integration of the disciplines of economic theory, mathematics, and statistics. It is the extremely useful and marketable skill though which applied economists earn a living, whether that be forecasting sales for a firm, forecasting the growth rate of GDP, or testing interesting hypotheses. Through the use of modern software and the explosion of data available on the web, it is extremely easy to download a data set and fit a line to it. The art of good econometrics is in understanding both the power and limitations of econometric models and in their careful interpretation. Consequently, this course is both a course in careful econometric theory and in the use of a modern econometric software package, STATA.

We begin with a review of basic statistical theory, building up to single variable regression equations under ideal conditions. Since conditions in the real world are almost never ideal, we study regression when the ideal conditions are not met

Course Learning Objectives: After successfully completing ECO 5347, you will be able to

1.  Identify common statistical distributions.

2.  Understand hypothesis testing as measuring the observed degree of divergence from the value you expect to find.

3.  Write STATA programs.

4.  Understand the full ideal conditions of OLS regression, and the properties of OLS estimators when one or more conditions are violated.

5.  Estimate logit, probit, and tobit models and understand when they are more appropriate than OLS.

6.  Estimate fixed effects and random effects models and understand when each is more desirable to use.

Administration

Required Course Materials:

Kmenta, Jan, Elements of Econometrics, 1986, Second Edition, McMillan

Baum, Christopher F., An Introduction to Modern Econometrics Using STATA, 2006, STATA Press

Journal Articles: All but one of the following journal articles are available at my personal web site, www.econjimwest.com . I will send the NBER working paper.

Does Drinking Impair College Performance? Evidence from a Regression Discontinuity Approach, Carrell, Scott E., Mark Hoekstra, and James E. West, Journal of Public Economics, 2011, 95(1-2), pp 54-62

Peer Effects in Academic Cheating, Carrell, Scott E., Frederick V. Malmstrom, and James E. West, Journal of Human Resources, 2008, 43(1), pp 173-207

Does Professor Quality Matter? Evidence from Random Assignment of Students to Professors, Carrell, Scott E. and James E. West, NBER Working Paper 14081, 2008

Does Professor Quality Matter? Evidence from Random Assignment of Students to Professors, Carrell, Scott E. and James E. West, Journal of Political Economy, 2010, 118(3), pp 409-432

Software: STATA/SE version 11 is available on the Hankamer School of Business server (H:\stata11\statase.exe), which you may use to complete all empirical assignments in this course. However, while you are a student, you have the opportunity to purchase your own copy of STATA at a sharply reduced price. If this is of interest to you, please contact me and I will set up a “STATA Grad Plan” which will allow you to purchase STATA at a student price.

Course Grading: The grade that you earn in the course will be based on your performance in each of the following categories shown weighted by the appropriate percentage:

Assessment / Percentage
Homework Assignments / 20
Writing Assignments / 20
Midterm Exam / 20
Final Exam / 40

Homework Assignments: We will have weekly homework assignments due on Wednesday. These will be a combination of theoretical exercises and brief STATA-based assignments.

Writing Assignments: For two of the journal articles covered in class, you will write a brief review of the econometric techniques used. You will explain what was to be tested, how the techniques used attempted to accomplish this, and other techniques that could or should have been used.

Midterm Exam: The midterm exam will be almost exclusively on theoretical content covered in the first half of the course.

Final Exam: The final exam will be held Tuesday, December 13 from 9-11am and will be cumulative. Since the final exam is your evidence of having mastered material covered in the course, failing the final exam will result in failing the course.

Late Work: Assignments are due at the beginning of class. If you will be absent, you must make arrangements to turn in your assignment at or before this time (note: This does not apply for unexpected illnesses, etc.). The clock continues through weekends and holidays. Late assignments will be penalized 1 letter grade per day late.

Attendance Policy: The Hankamer School of Business Attendance Policy states that you must attend a minimum of 75% of course lessons or will receive a final grade of F. A link to the policy and a more complete explanation of it is here. http://www.baylor.edu/business/economics/index.php?id=84087

Lesson / Topic / Assigned Reading / Major Assignments
Aug 22 / Introduction
Aug 24 / Inference
Sampling Distributions / Kmenta 1
Kmenta 2
Aug 29 / Probability and Distributions / Kmenta 3
Aug 31 / Sampling Distributions / Kmenta 4
Sept 7 / Tests of Hypotheses / Kmenta 5
Sept 12 / Estimation / Kmenta 6
Sept 14 / Working with data in STATA / Baum 2
Sept 19 / Simple OLS and Gauss-Markov / Kmenta 7.1 – 7.3
Baum 4.1 – 4.2
Sept 21 / Distribution of Estimators / Kmenta pp 225-243
Sept 26 / Hypothesis Testing / Kmenta pp 243-254
Sept 28 / Regression Output
Hypothesis Testing / Baum 4.3-4.6
Oct 3 / Review
Oct 5 / Midterm Exam
Oct 10 / Review Midterm Exam
Non-normality / Kmenta 8.1
Oct 12 / Heteroskedasticity - Theory / Kmenta pp 269-285
Oct 17 / Heteroskedsaticity - Practice / Kmenta pp 285-298
Baum 6.1 – 6.1.3
Oct 19 / Autocorrelation / Kmenta pp 298-320
Oct 24 / Moving Average / Kmenta pp 320-334
Oct 26 / Multiple Regression / Kmenta pp 392-412
Oct 31 / Hypothesis Testing / Kmenta pp 412-430
Nov 2 / Multicollinearity
Specification Errors / Kmenta 10.3
Kmenta 10.4
Nov 7 / Binary Regressors / Kmenta 11.1
Drinking
Nov 9 / Nonlinear Models / Kmenta 11.3 / Writing Assignment 1 due
Nov 14 / Qualitative Dependent Variables / Kmenta 11.5
Baum 10.1 – 10.2
Nov 16 / Limited Dependent Variables / Kmenta 11.6
Baum 10.3
Nov 21 / Simultaneity / Kmenta 13.1
Cheating
Nov 28 / Panel Data Methods / Handout
Baum 9.1
Nov 30 / Panel Data Continued / Professor Quality
NBER & QJE
Dec 5 / Review / Writing Assignment 2 due
Dec 13 / Final Exam
9-11am

3