st: xtnbreg Initial values

[Ivan Png <iplpng@gmail.com>]

Wonder if you can please help.

I want to run a panel fixed effect negative binomial regression.
  * dep variable: patentcount
  * indep variables: xrd revt capx yr1976-yr2006

I ran:
 xtnbreg patentcount xrd revt capx yr1976-yr2006 , fe

and got error message: "note: 395 groups (395 obs) dropped because of
only one obs per group
note: 2893 groups (20642 obs) dropped because of all zero outcomes

initial values not feasible
r(1400);"

I read somewhere that I should first run the specification in OLS and
use the OLS coefficients as starting values.  Accordingly, I ran:
xtnbreg patentcount xrd revt capx yr1976-yr2006 , fe ///
    from(xrd=0.0779 / revt=0.000316 / capx=-0.000964 ///
    yr1976=6.288 / yr1977=6.232 / yr1978=5.838 / yr1979=5.838 ///
    yr1980=6.074 / yr1981=5.793 / yr1982=6.322 / yr1983=5.568 ///
    yr1984=6.518 / yr1985=7.465 / yr1986=7.993 / yr1987=8.570 ///
    yr1988=9.873 / yr1989=10.825 / yr1990=11.180 / yr1991=11.793 ///
    yr1992=12.434 / yr1993=12.871 / yr1994=14.594 / yr1995=17.123 ///
    yr1996=17.334 / yr1997=20.174 / yr1998=19.188 / yr1999=20.503 ///
    yr2000=21.267 / yr2001=21.062 / yr2002=17.681 / yr2003=10.428 ///
    yr2004=1.522 / yr2005=-4.626 / yr2006=-8.280 )

Then I got the error message: "note: 395 groups (395 obs) dropped
because of only one obs per group
note: 2893 groups (20642 obs) dropped because of all zero outcomes
initial vector: matrix / not found
r(111);"

Can you please advise me?  How should I set up the xtnbreg?  What is
wrong with my syntax in specifying "from(...)"?

Thanks
Ivan Png

*
*   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/