Re: st: Fixed effects = insufficient observations?

[Tharshini Thangavelu <thth4658@student.su.se>]

I have already double checked the education variable and it is an integer
variable. I can do a OLS including edu without any problem
but when I do a fixed effects. I get a message: insufficient observations.

/ Tharshini

On Fri, 17 Sep 2010 13:55:09 +0200, Ulrich Kohler <kohler@wzb.eu> wrote:
> Am Freitag, den 17.09.2010, 11:59 +0200 schrieb Tharshini Thangavelu:
>> When I try to do a fixed effects regression, I get a message:
>> insufficient
>> obervations. I try with different specifications.
>> As my control variables, I have, all doingbusines variables(only
>> rankings), GDP per capita, education, unemployment and industry
>> structure.
>> When I exclude education variable (= gross enrollement rate (%)
secondary
>> from teh WorldBank, the obervations can be more than 100), it seems to
be
>> working. 
>> 
>> The education variable is a integer variable and I have other variables
>> that are integer without any kind of problems. Have someone had the
same
>> problem with this variable? 
> 
> 
> Are you sure that education is not a string variable holding integer
> numbers? Check out the storage type in output of -describe-. If the
> storage type is str## you must use -destring- first.
> 
> Uli
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> *
> *   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/

-- 
____________________________________

Tharshini Thangavelu

Forskarbacken 19, 2tr

114 16 Stockholm

Mob 0735 53 43 90

Epost thth4658@student.su.se
*
*   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/