This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Section 9  Econ 140 GSI: Hedvig, Tarso, Xiaoyu * 1 Panel Data Regression 1.1 Data types  Panel data • Reminder: Country Year GDP Interest rate Europe Brazil 2000 100 22 Brazil 2001 150 18 Brazil 2002 170 15 Brazil 2003 220 12 USA 2000 1000 4 USA 2001 1150 3 USA 2002 1400 2 USA 2003 1900 2 Italy 2000 60 6 1 Italy 2001 130 5 1 Italy 2002 150 5 1 Italy 2003 190 3 1 • 2 types: balanced : all variables are observed for each entity and each time period unbalanced: has some missing data for at least one time period for at least one entity 1.2 (Entity) Fixed E ects Regression Model • Regression model: Y it = β + β 1 X it + β 2 Z i + u it , where: i = 1 ,...,N t = 1 ,...,T Y it is the dependent variable X it are the explanatory variable u it is the error term Z i is an unobserved variable that varies with entities but does not change overtime. Example : Suppose that the entity is a state. Then Z i can represents cultural attitudes toward drinking and driving for the state i . * Many thanks to previous GSIs, Edson Severnini and Raymundo M. CamposVazquez, as this note is based on theirs. All errors are ours. 1 • The regression model above can be rewritten as Y it = α i + β 1 X it + u it , where α i = β + β 2 Z i . This equation is the (entity) xed e ects regression model ....
View
Full Document
 Spring '08
 DUNCAN
 Econometrics, Regression Analysis, Xit, UIT, Yit, Fixed Eects Regression

Click to edit the document details