Re: st: xtset and xtreg for panel data

[Jeph Herrin <stata@spandrel.net>]

Sounds like you want to use -xtmixed- so you can model the variance appropriately.
Something like

 xtmixed depvar indvars || country: || station:

would account for multiple obs at both station and country level.

hope this helps,
Jeph


On 11/27/2011 3:08 PM, Andrew Wong Yoon Loong wrote:
> Hi Statalisters,
>
> I am working with panel data that is similar to the following:
>
> Station   Country        Year    Var1   Var2
> 1              US            1990    25      25
> 1              US            1991    15      18
> 2              US            1990    60      12
> 2              US            1991    300     50
> 3              BR            1990    25      25
> 3              BR            1991    300     50
> 4              BR            1990    25      25
> 4              BR            1991    300     50
>
> I'm trying to use the xtreg command to perform both fe and re models.
> However, I can't perform the initial xtset country year because of
> duplicates. Is there a way around this problem? I know one option is
> simply to employ LSDV and create a dummy for each country but I'm
> hoping to still be able to use xtreg. Thanks in advance for your help.
>
> Regards,
> Andrew
>
> *
> *   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/
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2012.0.1873 / Virus Database: 2101/4642 - Release Date: 11/27/11
*
*   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/