RE: st: Dynamic Panel Data

[DE SOUZA Eric <eric.de_souza@coleurope.eu>]

First, let's ignore the fact that your data are five year averages, and consider them as successive time periods. In this case, the introduction of a lagged value of the dependent variable as regressor makes FE, RE and FD inconsistent because the lagged dependent variable, when transformed in order to apply the above methods are correlated with the residuals. You, therefore, have to use instrumental variables. LDV, because it is identical in effects to FE, has the same problem. Stata has a dynamic panel data routine but I have never used it. What I use is -xtabond2- written by David Roodman and which can be downloaded using -ssc install xtabond2- . The method is not simple and would require some reading up on your part. But if you Google, you will find quite a lot of course notes explaining it. You also have the paper by David Roodman himself, but I personally do not recommend it for beginners. If you wish to use the built-in Stata commands , they are -xtabond- and -xtdpdsys!
 - .

You can find an excellent introduction by Stephen Bond here:
http://cemmap.ifs.org.uk/wps/cwp0209.pdf

You can also find a set of lecture notes here:
http://www.nuff.ox.ac.uk/users/bond/
under CEMFI summer school

A good discussion of panel data models can be found in chapters 10 and 11 of J. Wooldridge, Econometric Analysis of Cross Section and Panel Data, second edition, 2010

Does the fact that you use five averages affect the results? I cannot answer that. May be Jeffrey Wooldridge may intervene here.


Eric de Souza
College of Europe
Brugge (Bruges), Belgium
http://www.coleurope.eu


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Humaira Asad
Sent: 06 March 2011 00:53
To: STATA HELP
Subject: <SPAM>st: Dynamic Panel Data



Hi,
 
I am working with a panel of 107 countries spanning over 1960 to 2010. Since there were a large number of missing values so I have averaged the data over five years. Now there are 10 time periods. I have estimated the models using OLS, Fixed Effect and Random effect. The results show that random effect model is appropriate. When I introduce the lags of the dependent variable, estimates considerably change and very few regressors remain significant. It seems I am wrong somewhere.
 
1. How can I decide the number of lags of the dependent variable? as the first lag of the dependent variable appear to be significant, but its inclusion makes almost all the other regressors insignificant.
2. Is it econometrically correct to use LDV when data is averaged over five years?
3. Kindly suggest any reading to be clear about these things?






Humaira Asad
PhD Research Scholar

UoE Business School

University of Exeter, England 		 	   		  
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/