Panel Data Analysis using STATA

Course objectives
This course gives an overview of a selection of key topics in panel data econometrics, a branch
dealing with data structures which have both time and group dimensions. Although the emphasis
of the course is on practical implementation of the techniques to be learned, a sound knowledge of
statistics, basic econometrics and matrix algebra will help make the most of the course material. The
techniques to be learned during the course will be demonstrated using STATA statistical software,
a good command of which is therefore a bonus.

Course content

• 1 Introduction
• Examples of panel data.
• Why use panel data?
• The N and T dimensions.
• Declaring a panel dataset in STATA: xtset command.
  
  2 Basic regressions with error components
• Fixed effects.
• Random effects.
• STATA's xtreg command.
• Fixed or random effects?
• hausman command.
• Fixed and random effects reconciled: The Mundlak and Hausman-Taylor estimators.
  xthtaylor command.
  
  3 Hypothesis testing with panel data
• Testing for and addressing residual heteroscedasticity.
• xtgls command.
• Testing for serial correlation.
• xtserial command.
• Addressing residual correlation: xtpcse command.
  
  4 Dynamic panel data
• The problem: bias in the estimates for short T.
• An overview of GMM estimators: Arellano and Bond, Arellano and Bover and Blundell and Bond system estimator.
• xtabond and xtdpdsys commands.
  
  5 Special topics
• The logit regressions on panel data.
• xtlogit command.
• Count data regressions.
• xtpoisson command.

Participants profile
The course is open to graduate students, postdoctoral fellows as well as practising researchers.
Prior to the course, participants should have working knowledge of regression models and practical
knowledge of a statistical packages such as STATA, SPSS, EViews, R,  ... ezv. The course is open to a maximum of 20 participants.

Schedule
The course  is scheduled from 9:00am to 4:00pm, March,  2012. Exact date will be announced

Course fee
Course fee: 1325 € (Euro), VAT-free

PhD sponsorship
Research centers, government organizations as well as universities can sponsor one or more PhD candidate to participate to this course  

Location:
Laan Copes van Cattenburch 62, The Hague

Language
Although both lecturers speak Dutch, English language is the working language of the course.

Teaching team
This course is alternately given by dr. N. Zubanov & dr. S. Vujić

Readings
Microeconometrics Using Stata, Revised Edition by A. Colin Cameron, Pravin Trivedi

Dr. S. Vujić
Dr Vujić is a Senior Researcher at the London School of Economics. She received her Ph.D. degree from the Economics and Econometrics Departments of the Free University in Amsterdam. Her main area of interest is quantitative research on socio-economic questions, in the fields of economics of crime, economics of education, behavioural economics, and health economics, which involve use of large panel and survey data sets. Her other work experience includes posts at the Centre for Economic Performance (CEP) at the London School of Economics, The Netherlands Bureau for Economic Policy Analysis (CPB), and ORTEC Finance. She also held lectureship positions at various universities: Free University Amsterdam (statistics, economics and econometrics), University of Freiburg (economics of education), University of Groningen (microeconometrics and statistics). She is affiliated with Tridata as a lecturer for STATA software, with applications to survival and panel data analysis.

Dr. N. Zubanov
Dr. Zubanov received his PhD in applied labour economics from Birmingham Business School in 2007. After having worked for 2.5 years at CPB Netherlands Bureau of Economic Policy Analysis he took up an assistant professor position at the University of Tilburg. He does research and publishes in the areas of personnel and labour economics, where he gained plenty of experience in working with large panel datasets. Dr. Zubanov is affiliated to Tridata as a consultant and lecturer for STATA software, with applications to panel data analysis.

Application
 Please fill in the application form located here http://www.tridata.nl/folder/aanmeldingsformulier.pdf
Application can either be sent by mail or email.

Application form