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: st: Panel cointegration

From   Nick Cox <>
Subject   Re: st: Panel cointegration
Date   Thu, 25 Aug 2011 07:46:42 +0100

Whether any participant is confused or not, this is confusing to
onlookers. -xtpmg- is user-written and you are asked to explain where
user-written software you refer to comes from.

SJ-7-2  st0125  . . . . . . . Estimation of nonstationary heterogeneous panels
        (help xtpmg if installed) . . .  E. F. Blackburne, III and M. W. Frank
        Q2/07   SJ 7(2):197--208
        estimates nonstationary heterogeneous panels in which the
        number of groups and number of time-series observations
        are both large

Further, if either of you has read anything by Pesaran on this, a full
reference to literature or other source is expected.

Please do read the Statalist FAQ for reminders of what is requested in


On Thu, Aug 25, 2011 at 2:15 AM, Manhal Mohammad Ali
<> wrote:
> I think what Pesaran is saying by large (or maybe long) panel is when
> you have large T observations but small N.
> On Thu, Aug 25, 2011 at 1:16 AM, dar mark <> wrote:

>> Thank you for the explanation Biljana. I am just concerned about what
>> Pesaran says that xtpmg is for large N large T panels and mine is
>> relatively small. But at the same time he has applied it on small N
>> small T panels. Therefore I am litlle bit confused.

On Wed, Aug 24, 2011 at 8:34 PM, Biljana Jonoska
<> wrote:

>>>> I used xtpmg which produces pmg and mg estimates at the level of panel. PMG estimator is an ARDL based and it is special case of the VAR model. You can use PMG estimator only if there is no need of examination of simultaneous interdependence among the variables.Namely, ARDL regression is designed to establish ECM in the case when there is only one-way interdependence between dependent and explanatory variables at the level of
>>>>  cross-sections in panel, whereas PMG does the same as ARDL at the level of panel.
>>>> STATA offers basic help for usage of xtpmg.ado package.

On Tue, 8/23/11, dar mark <> wrote:

>>>> I want to do a panel cointegration analysis for N=5 T=20 panel. I have
>>>> searched for estimators and found Pesaran's xtpmg. Can anyone tell me
>>>> can I apply it and if not suggest another
>>>>  estimator?

*   For searches and help try:

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