Josť Fajardo


Panel Data Analysis

Objective

The aim of this course is to present the background necessary to understand and assess the applications of panel data analysis and to provide skills which could be applied to analyse a variety of research and policy problems related to Business. Both static and dynamic models for panel data analysis are presented, with special attention to choosing the most suitable estimator for each model. In particular, the course covers the different structures of data and the advantages and limitations of panel data. Moreover, several cases on how to derive economic models combining panel data and Stata are discussed.

Course Outline

Linear Static Models for Panel Data  and  Nonlinear Panel Data Models

Slides: class0, class1,class2

Data Sets:  Crime2, Kielmc, CR, Beatles, MarriagePremium,

Do_files:  class1, class2

(obs: First save the .do and .dta files and then open in Stata. In .do files correct the path directory where you have downloaded the data)

Problems Set:

Reading List:

Date to be Presented  and Papers

23/01 H&M, MarriagePremium

30/01 Happiness&Age, AERHyp

Grading

        Problem set assignments and Readings: 30%

        Mid term evaluation: 30%

        Research Paper presentation: 40%

Bibliografia

        Arellano, M. (2003), Panel Data Econometrics (Advanced Texts in Econometrics), Oxford University Press.

        Bond, S.R. (2002): “Dynamic Panel Data Models: A Guide to Micro Data Methods and Practice”. Cemmap Working Paper Series No. CWP09/02, Institute for Fiscal Studies, London.

        Greene, W. (2000): Econometric Analysis (5th Edition), Prentice Hall

        Roodman, D. (2006): “How to Do xtabond2: An Introduction to “Difference” and “System” GMM in Stata”. Working Paper 13, Centre for Global Development.

        Wooldridge, J. M. (2001), Econometric Analysis of Cross Section and Panel Data, MIT Press.

        Cameron and Trivendi (2010), Microeconometrics Using Stata. STATA Press.

Additional Reading: GreenePanelData, Stata_PanelData, xtdpd, Time_Series

 


back

home