st: re: time dummies

[Kit Baum <kitbaum@mac.com>]

Grant said

I am estimating my panel data model using xtabond2 with time dummies  
as follows;
yr70_74, yr75_79, yr80_84, yr85_89, yr90_94, yr95_99, yr00_04. Thus,  
my T=7 ( where t=1...7). However, before including the time dummies  
the model starts estimating at period 3. Does it mean that I can only  
include my time dummies starting with yr80_84 or yr75_79?


It is standard with Arellano-Bond difference estimator to lose the  
first two obs. of your level series. It is a model in first  
differences, so you lose one obs. defining the dependent variable,  
and as the model is dynamic you include the lagged difference as a  
regressor. Defining the lagged difference of a level requires two  
observations.

Through sleight of hand, the GMM-System estimator of Blundell-Bond  
(which is in fact the default in xtabond2) only loses one  
observation, as it can be thought of in levels with a lagged level as  
a regressor. Thus with GMM-System you should be able to include time  
periods 2..7.

Kit

Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/