Dear Statalisters,

may be someone can help me with the following (hopefully small and easy) problem:

I have a panel with 26 panel units and a time variable from 1 - 19 ( 1984 to 2001, respectively).
I have tsset-ed the data and tried to run the following regression

newey logKrimrate1a ........... year1986 - year1997 if year > 1985 & year < 1999 , lag(1)

which resulted in the message:

"year is not regularly spaced"

There are some missing values in my dataset, but not for the time period under investigation (1986 - 1998). year1986 - year1997 are time dummies.
This message occurs independently of if I use a year variable (1984 - 2001) or a time period variable running from 1 to 19.
If I use newey2 ( instead of newey), however, a regression result is displayed.

So my question is, why does this message above occur ????
Or is 'newey' inappropriate for panel data ?


Yours sincerely,

Justina

Dipl.Vw. Justina A.V. Fischer,
Research Associate
SIAW - University of St. Gallen
Dufourstr. 48
CH-9000 St. Gallen
Tel.:++41-71-224 2345
Fax: ++41-71-224 2298
www.siaw.unisg.ch