Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: GMM or 2SLS with cluster adjustment or others


From   "Rodrigo Alfaro A." <ralfaro@bcentral.cl>
To   "Kit Baum" <baum@bc.edu>, "Yu-Luen Ma" <yma@ilstu.edu>
Subject   st: RE: GMM or 2SLS with cluster adjustment or others
Date   Thu, 15 May 2008 09:11:17 -0400

Kit and Yuluen,

I would not recommend using Arellano-Bond (AB) routinely too. For a small T (number of time periods), AB is fine, whatever is the theoretical model behind. 

For moderate or large T, I suggest to truncate the number of lags used in AB (call this estimator TAB) or you can use LDV+FE instead. The bias of the latter decreases on T (see note 1). Also, TAB is almost unbiased, but it is less efficient than AB (see note 2). 

In conclusion, I suggest to start with a estimator that agrees your theoretical model as first choice. If there is "no-model" but endogeneity then you can use IV estimators as Kit suggested.

Best regards, 
Rodrigo.

Notes: 
(1) The early paper on the bias of LDV+FE is Nickell, S. (1981) "Biases in Dynamic Models with Fixed Effeects," Econometrica, 49, 1417-1426. For moderate or large T, you have Hahn, J., and Kuersteiner, G. (2002) "Asymptotically Unbiased Inference for a Dynamic Panel Model with Fixed Effects When Both n and T are Large," Econometrica, 70,
1639-1657 and Alvarez, J., and Arellano, M. (2003) "The Time Series and Cross-Section Asymptotics of Dynamic Panel Data," Econometrica, 71, 1121-1159. A more friendly paper is Beck, N., and Katz, J. (2004) "Time-Series-Cross-Section Issues: Dynamics," Working Paper, New York University, probably published in some Political Science journal. A good survey on Panel Danel that deals this material is Arellano, M. (2003) Panel Data Econometrics, New York: Oxford University Press.

(2) A practical issue on AB when T is moderate or large is to truncate the number of lags used as instruments. I found that this practical procedure reduces the finite-sample bias of AB (see pages 7/8 and the appendix A.2 of my WP available at http://www.bcentral.cl/eng/studies/working-papers/pdf/dtbc467.pdf). However, Monte Carlo simulations (not reported) show that the TAB is less efficient than AB, then there is a trade-off between finite-sample bias and efficiency. That could be solved by finding the optimum lag that miminizes the MSE. I am working (in my sparse time = low priority) on that, but it will require a balanced panel.


 

-----Mensaje original-----
De: Kit Baum [mailto:baum@bc.edu] 
Enviado el: Miércoles, 14 de Mayo de 2008 07:11 p.m.
Para: Rodrigo Alfaro A.
CC: statalist@hsphsun2.harvard.edu
Asunto: Re: GMM or 2SLS with cluster adjustment or others

Rodrigo,

The point I was trying to make is that the fixed effects estimator (e.g. xtivreg, fe) is biased in the presence of a LDV, and so Arellano-Bond should be used in that context. There is nothing to prevent you from using Arellano-Bond in an equation without a LDV, but given the complexity of the A-B methodology (and possibility to generate hundreds of instruments) I would not recommend using it routinely on a static panel data model. In that context it is certainly appropriate to use IV-GMM, as is provided by xtivreg2, fe with the gmm2s option. That is also a GMM estimator which will be more efficient than standard IV.

Best wishes
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


On May 14, 2008, at 18:39 , Rodrigo Alfaro A. wrote:

>
> Kit wrote:
> The Arellano-Bond / Arellano-Bover / Blundell-Bond methodology 
> (xtabond,
> xtabond2) is appropriate if you have lagged dependent variables, and 
> unnecessary otherwise.
>
> I am not agree with this statement. Equation (13) in the classic paper 
> Blundell, Bond, Devereux, and Schiantarelli (1992) "Investment and 
> Tobin's Q: Evidence from company panel data" Journal of Econometrics 
> 51
> (1992) 233-257 does not have a lagged dependent variable, but it is 
> also estimated by GMM.
>
> Rodrigo.
>
> **********************************************************************
> **********
> ADVERTENCIA: La  información  contenida  en  esta  transmisión, y   
> en  cualquier archivo  adjunto, está  sujeta a reserva legal conforme 
> a la normativa aplicable  al  Banco  Central  de  Chile, y  no  puede  
> ser usada o difundida  por personas distintas  de  su o sus 
> destinatarios. Si usted ha recibido esta transmisión por error,  por  
> favor  notifique  inmediatamente al remitente respondiendo por este 
> mismo medio y elimínela de su sistema.
> El  Banco Central de Chile no se hará responsable de la exactitud y  
> veracidad de la información contenida en este mensaje, así  como   
> de su  modificación, copia, divulgación  o  reenvío,  total  o   
> parcial.   Su  uso  no  autorizado puede ser sancionado de  
> conformidad con las leyes chilenas.
> El  Banco  Central  de  Chile  transmite  sus decisiones a través de 
> comunicados oficiales, los  que  pone  a  disposición  del público en 
> su página de Internet: www.bcentral.cl
>
>
> DISCLAIMER: The information  contained  in  this  email or any 
> attached file, is subject to legal  privilege  pursuant  to the laws 
> and regulations applicable to the Central  Bank  of  Chile ,
> and may not be used or disseminated by any person other  than  its   
> intended recipients. If you have received this transmission in error, 
> please  notify  the sender immediately by reply to this email address 
> and delete it from your system.
> The Central Bank  of  Chile shall not be liable for the accuracy or 
> authenticity of the contents of this message, whether amended, copied, 
> forwarded or disclosed in  any  form, in  whole  or  in
> part.  Please note that unauthorized use may be penalized  in   
> conformity  with  the  Chilean law.
> The Central  Bank of Chile communicates its decisions by  official 
> releases, and makes them available to the public in its WebPages: 
> www.bcentral.cl


********************************************************************************
ADVERTENCIA: La  información  contenida  en  esta  transmisión, y  en  cualquier archivo  adjunto, está  sujeta a reserva legal conforme a la normativa aplicable  al  Banco  Central  de  Chile, y  no  puede  ser usada o difundida  por personas distintas  de  su o sus destinatarios. Si usted ha recibido esta transmisión por error,  por  favor  notifique  inmediatamente al remitente respondiendo por este mismo medio y elimínela de su sistema.
El  Banco Central de Chile no se hará responsable de la exactitud y veracidad de la información contenida en este mensaje, así  como  de su  modificación, copia, divulgación  o  reenvío,  total  o  parcial.   Su  uso  no  autorizado puede ser sancionado de conformidad con las leyes chilenas. 
El  Banco  Central  de  Chile  transmite  sus decisiones a través de comunicados oficiales, los  que  pone  a  disposición  del público en su página de Internet: www.bcentral.cl 

DISCLAIMER: The information  contained  in  this  email or any attached file, is subject to legal  privilege  pursuant  to the laws and regulations applicable to the Central  Bank  of  Chile , and may not be used or disseminated by any person other  than  its  intended recipients. If you have received this transmission in error, please  notify  the sender immediately by reply to this email address and delete it from your system.
The Central Bank  of  Chile shall not be liable for the accuracy or authenticity of the contents of this message, whether amended, copied, forwarded or disclosed in  any  form, in  whole  or  in part.  Please note that unauthorized use may be penalized  in  conformity  with  the  Chilean law.    
The Central  Bank of Chile communicates its decisions by  official releases, and 
makes them available to the public in its WebPages: www.bcentral.cl

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index