Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st:AR(1) error notice: sample may not include multiple panels


From   Richard Herron <richard.c.herron@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st:AR(1) error notice: sample may not include multiple panels
Date   Sat, 17 Mar 2012 21:13:21 -0400

This question came up yesterday, too. Here was my response:

Panels require strict exogeneity for consistency, so you can't have
lagged dependent variables on the right hand side (i.e., dynamic
panels aren't allowed).

If you think about the within estimator for panel data, then you have
y_{i, t-1} - mean(y_i) on the right hand side, which is not orthogonal
to e_{i,t} because mean(y_i) contains y_{i,t} from the left hand side.

There are some solutions to this in -xtabond-, which does the Arellano
and Bond estimator. Arellano and Bond correct the endogeneity by
estimating the dynamic panel in first differences and using lags as
instruments and GMM to take care of over-identification.

Chapter 9 in Cameron and Trivedi's "Microeconometrics Using Stata"
provides a handful of techniques for estimating dynamic panel models.
Chapters 21 and 22 in their Microeconometrics textbook provides more
theory. HTH.

On Sat, Mar 17, 2012 at 18:02, Amanda Fu <mandy.fu1@gmail.com> wrote:
> Dear Statalists,
>
> Sorry for bothering you with  a question about a simple AR(1)
> estimation. I searched related discussion in the archives , but still
> cannot figure out why the error message comes out. Any suggestions
> will be helpful.
>
> I want to estimate a univarate AR(1) model (without any controls). The
> data set is a panel data,  including 1000 observations for 5 years
> (200 variables. The one I am interested is enroll taking values 0,1) .
> . tsset id years
> . arima enrollment,ar(1)
> "sample may not include multiple panels"
>
> I must have missed something here. Can someone give me some hints?
> In addition, instead of using ARIMA, may I know if there is any easy
> alternative to do the AR(1) ?
>
> Thank you very much!
>
> Amanda
> *
> *   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/


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