unit root test for panel data

[yan yan]

Dear all,
I'd like to run a unit root test for a panel data. The data is unbalanced. The panel variable(id) is 9976(max) and time variable(year) is 9.
I tried to run the ADF test but it is said in red "sample may not include multiple panels". Also, LLC is said not suitable for unbalanced data.
What I'm trying to do now is: use "gen t=_n" command to treat the dataset as a time series and then use ADF to test the unit root. Correct me if I'm wrong.
what's more, is it necessary to test the unit root for a panel dataset? Thanks for your time.

[Diana Hechavarria]

I have a similar problem. I am new to time series data as well. I am trying to complete the dfuller2 test (which I was told could be applied when dealing with multiple panels) and I get the following eroor "sample may not include multiple panels r(459);" --I have been searching for information on the boards and faq in STATA and I have not found a solution. Any help would be appreciated. Thank you for your time and expertise in advance.

[Carlo Lazzaro]

Yan and Diana:
I'm totally unfamiliar with this stuff, but following the guidance provided by http://www.stata.com/bookstore/micro...ata/index.html (pages: 278-279), the following user written commands may be interesting:
levinlin from http://fmwww.bc.edu/RePEc/bocode/l
ipshin from http://fmwww.bc.edu/RePEc/bocode/i
xtpmg from http://fmwww.bc.edu/RePEc/bocode/x

[Eric de Souza]

Please provide the exact instructions both of you typed.
Yan yan: with only 9 periods it makes no sense to test for unit roots.

[Diana Hechavarria]

I actually was able to solve my issue using the xtabond2 command. Thank you everyone for your help.