# st: Re: RE: Chow test in panel data

 From "Andrea Molinari" <[email protected]> To <[email protected]> Subject st: Re: RE: Chow test in panel data Date Fri, 2 Jul 2004 17:16:34 +0100

```Scott,
Thanks very much for your suggestion. I'm dividing the sample into three
subperiods because it would make (economic) sense to have different
coefficients among the periods.
What I did so far is a Chow which could have been done for time-series too,
and what I was mostly concerned was whether there is any test (Chow or
other) for this in a panel data context...
Andrea

----- Original Message -----
From: "Scott Merryman" <[email protected]>
To: <[email protected]>
Sent: Friday, July 02, 2004 5:03 PM
Subject: st: RE: Chow test in panel data

> Andrea,
>
> It might be easier help if you explained what you did to generate the 3
> groups and append them.
>
> Here is one example for testing that the coefficient vectors are the same
in
> two periods (this is Greene's Econometric Analysis 4th ed., page 289,
> Example 7.8).
>
> use http://fmwww.bc.edu/ec-p/data/Greene2000/TBL6-1, clear
>
> gen lng = ln(g/pop)
> gen lny = ln(y)
> gen lnpg = ln(pg)
> gen lnpnc = ln(pnc)
> gen lnpuc = ln(puc)
> mark dummy if year < 1974
> foreach var of varlist lny-lnpuc year {
>  gen `var'1 = dummy*`var'
>  }
> reg lng lny lnpg  lnpnc lnpuc year lny1- year1 dummy
> testparm lny1  lnpg1 lnpnc1 lnpuc1 year1 dummy
>
> Hope this helps,
> Scott
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Andrea Molinari
> Sent: Thursday, June 24, 2004 9:48 AM
> To: [email protected]
> Subject: st: Chow test in panel data
>
> Dear Statalisters,
> I am trying to run a Chow test to see if I can separate the 22 years
period
> into three subperiods. I have an unbalanced panel.
> I tried the suggestions made in the Stata website (FAQs):
> http://www.stata.com/support/faqs/stat/chow3.html
> but this doesn't seem to work in my case, since when I append the three
> subperiods the variables get missing values for those years in which there
> is no information.
> Has anyone got a suggestion on how best to do this?
> Thanks very much!
> Andrea
>
>
> *
> *   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/
>

*
*   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/
```