Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Re: Re: st: handling serial correlation in fixed effects estimations

From   Christopher Baum <>
To   "" <>
Subject   Re: Re: Re: st: handling serial correlation in fixed effects estimations
Date   Sun, 1 Dec 2013 14:46:55 +0000

On Dec 1, 2013, at 2:33 AM, Philipp wrote:

> Regarding unit roots: my dependent variable is the current account balance as a ratio of GDP and according to the literature, there might (!) be a unit root but it seems not to be clear for the reasons you gave (difficult to proove 100%). IF there is indeed a unit root I shouldn't use fixed effects, you said. What about pooled OLS? If I subdivide my T = 25 in, say, 5 x T = 5 periods with averages of the yearly values and then use pooled OLS, this unit root problem doesn't bother me anymore, correct?
> If I assume for now there is no unit root and I use fixed effects:
> Regarding autocorrelation: the way I test autocorrelation (by predicting residuals and then estimating "reg resid l.resid") doesn't work properly since autocorrelation differs across panels, is that what you mean? What would be an appropriate way to test autocorrelation?
> I already tested a lot of model varieties including models with lagged values of the explanatory variables, but the autocorrelation persists. So there is no way for me to get rid of it by specifying another model, I think. It only goes away when I use a lagged dependent variable or estimate the equation in first differences, but I'd like to avoid both options. Is there any other way to deal with autocorrelation? 		 	   		  
> *
Pooled OLS can almost always be rejected by fixed effects via the F-test at the foot of the output. If it rejects, pooled OLS is inconsistent.

If you find that a lagged dependent variable is significant, then the observed autocorrelation is almost surely an indication that the static specification is inadequate, and the results are biased and inconsistent. I would explore a DPD/Arellano-Bond model (e.g., using Roodman's xtabond2 from SSC) in this context. It provides the A-B "abar" tests of autocorrelation; in difference GMM, you expect significant negative AR(1) by construction, but should not find significant AR(2).


Kit Baum
Professor of Economics and Social Work, Boston College, Chestnut Hill MA, USA
DIW Research Professor, Department of Macroeconomics, DIW Berlin, Berlin, Germany  |

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index