In the first part of this course, we will examine the relationship between education and earnings using earnings regressions and data from the 2001 Current Population Survey (CPS) March Demographic Supplement.
READINGS
The required textbooks for this course are:
Wooldridge, Jeffrey, Introductory Econometrics: A Modern Approach, South-Western College Publishing (Thomson Learning), andFor the basic theoretical and econometric models, as well as a review of the evidence, regarding the relationship between earnings and education, we will read from:Peter Kennedy,1998, A Guide to Econometrics, fourth edition, The MIT Press, Cambridge.
Ernst Berndt, 1991, The Practice of Econometrics: Classic and Contemporary, Addison-Wesley, New York, chapters 5 and 11.In addition, there will be other short readings from the CPS web site, journals and the NBER working paper series. These are on the reading list.
GRADING
Grading for part 1 of the course will be based on 5 problem sets, one take-home mid-term exam, and class participation.
Each problem set will be worth 12% of your grade for part 1 of the course. You may collaborate on the problem sets with other students, but you must turn in your own copy of the completed problem set, and you must do your own computer work (for example, you must run your own STATA programs). You must hand a hard copy of your answers to each problem set to me, as well as a hard copy of the STATA log files that show the work you did to answer the problem sets.
The mid-term exam will be worth 20% of your grade for part 1 of the course. The mid-term will be similar to a problem set except that you must do all of the work on the mid-term yourself (you cannot collaborate with anyone else).
Class participation will count for the remaining 20% of your grade for part 1 of the course.
I. The Current Population Survey (January 30 - February 6)READING LIST
ECONOMETRIC METHODS II, PART 1
UMBC ECONOMIC POLICY ANALYSIS PROGRAM
Almost all of the information you will need about the Current Population Surveys is on a web site, http://www.bls.census.gov/cps/. At a minimum, you should read the following:
To understand the Basic CPS survey, click on “Basic Monthly Survey” and read the Description and History sections. Also, click on Methodology and Documentation and read the sections on: methodology overview, sampling, weighting, data collection description, and producing summary statistics. Under the section titled "reviewing and revising the data," read the description and the section on topcoding of earnings.
To understand the March Supplement, click on the box marked “Supplements” and then on “Annual Demographic Survey (March CPS Supplement)” and read the Description and History sections. Click on the 2001 Methodology and Documentation and read the sections on: methodology overview, weighting, sampling and the description of the data collection. From the March Supplement home page, click on the 2001 Data to get to FERRET.
The variables available in the March CPS public use data sets are listed and described in the Glossary of Subject Concepts and in the Data Dictionary. The questionnaire and the interviewer’s manual for the basic CPS survey are available under Basic Monthly Survey, Methodology and Documentation, Data Collection. The questionnaire and the interviewer’s manual for the March Supplement are available under 1996 Data, Methodology and Documentation, Data Collection. Documentation is also available through FERRET.
Recommended reading:
Jin Heum Park, 1996, "Measuring Education Over Time: A Comparison of
the Old and New Measures of Education from the Current Population Survey,"
Economic Letters, Vol. 50, pp. 425-428.
Harley Frazis and Jay Stewart, 1999, "Tracking the Returns to Education
in the 1990s; Bridging the Gap Between the New and Old Current Population
Survey Education Items," The Journal of Human Resources, Vol. 34, No.3,
pp. 629-641.
II. Introduction to STATA (February 11-13)
STATA User's Guide
III. Estimating Earnings Equations and Returns to Education (February 18-20)
Berndt, Chapter 5, "Analyzing Determinants of Wages and Measuring Wage
Discrimination: Dummy Variables in Regression Models," pages 150-174.
IV. Some Econometric Issues in the Estimation of Earnings Equations and the Returns to Education (February 25-March 4)
Omitted Variables:
Wooldridge, 87-92.
Kennedy, 94-95, 101-104.
Orley Ashenfelter and Cecilia Rouse, 1999, "Schooling, Intelligence,
and Income in America: Cracks in the Bell Curve," NBER Working Paper 6902,
January. (This working paper is available at http://www.nber.org/
.)
Multicollinearity:
Wooldridge, 92-102.
Kennedy, Chapter 11.
Heteroskedasticity:
Wooldridge, 248-261.
Kennedy, pages 119-121, 133-135.
Specification:
Wooldridge, 278-291.
Kennedy, Chapters 5 and 6.
Weighted Regressions using Survey Data:
William DuMouchel and Greg Duncan, 1983, "Using Sample Survey Weights
in Multiple Regression Analysis of Stratified Samples," Journal of the
American Statistical Association, Vol. 78, No. 383, September, pages 535-543.
Recommended reading for the analysis of clustered and stratified surveys:
Angus Deaton, 1997, The Analysis of Household Surveys, Johns Hopkins
University Press for the World Bank, Chapters 1 and 2, Baltimore.
Stata User's Guide, Chapter 30 (Overview of Survey Estimation).
V. Qualitative Dependent Variables (Probit and Logit) and Correcting for Selectivity Bias (March 6-13)
Berndt, Chapter 11, "Whether and How Much Women Work for Pay: Application
of Limited-Dependent Variable Procedures," pages 593-610.
Wooldridge, 529-540, 551-564.
Kennedy, Chapter 15 and pages 251-252, 256-257, 258-259.