Bookmark and Share

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


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

st: Re: xtserial


From   Christopher Baum <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: Re: xtserial
Date   Sat, 16 Oct 2010 07:09:28 -0400

<>
On Oct 16, 2010, at 2:33 AM, Amin wrote:

> I have used -xtserial- and the results suggest that there is NO first
> order autocorrelation in my error terms. (all the independent
> variables are in current time same as dependent variable)
> 
> Based on this result, can I simply assume there is no need to embark
> on a dynamic panel data analysis, and a fixed effect analysis would be
> sufficient?

I don't see the connection. The absence of autocorrelation in a panel model does not imply that its dynamics are specified properly. In writing a static model (all variables at time t), you are presuming that the full effects of the Xs are represented by the contemporaneous value of y. That is sometimes a sensible assumption, but in many contexts ruling out dynamics is a bad idea. I would try specifying this model in the DPD framework (i.e., xtabond, xtabond2) and examining the coefficient on the LDV. If it is insignificant, then the static model is justified.

Kit

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.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/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


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