Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# Re: st: xtnbreg - robusteness check and model relevance

 From Mário Marques To statalist@hsphsun2.harvard.edu Subject Re: st: xtnbreg - robusteness check and model relevance Date Fri, 18 Jan 2013 10:16:07 +0000

```Dear Verkulien,

Thank you for your advising lr test, I have performed that,as also
Andre suggested, and it works.
Regarding the observations dropped because zero outcomes, it was not a
huge concern however it represents about 30% of country-pairs. If I
perform xtnbreg, re, an excerpt of the output is as follows. Do you
think it is plausible and convincing use a random effects instead?
I take the oportunity to ask you whether there is any way, under
xtnbreg, fe,re model, to obtain clustered/robust standard errors?

I would appreciate for any help. Thansk you.
Mário

-------------excerpt of the outcome ---------------
Random-effects negative binomial regression	Number of obs      =	11520
Group variable: country_pairs	Number of groups   =	1152

Wald chi2(17)      =	947.69
Log likelihood  = -16324.972	Prob > chi2        =	0.0000

nsubs       Coef.   Std. Err.      z	P>z     [95% Conf.	Interval]

x1    .4681317   .1032329     4.53	0.000      .265799	.6704644
x2   -.7630913   .2701788    -2.82	0.005    -1.292632	-.2335507
x3    .2865468   .0253246    11.31	0.000     .2369114	.3361822
x4    .7518349   .4471756     1.68	0.093    -.1246132	1.628283
x5    .0285722   .0385551     0.74	0.459    -.0469944	.1041388
x6   -1.452807   .3164543    -4.59	0.000    -2.073045	-.8325676
x7   -.0026751   .0027837    -0.96	0.337    -.0081312	.0027809
x8   -.3470533   .1021445    -3.40	0.001    -.5472528	-.1468537

year1...
year2...
cons   -1.704515   1.053902    -1.62	0.106    -3.770125	.3610946

/ln_r    .1649259   .0564414	.0543028	.275549
/ln_s   -1.255421   .0451007	-1.343817	-1.167026

r    1.179306   .0665617	1.055804	1.317254
s    .2849558   .0128517	.2608481	.3112914

----------------end of the excerpt-----------------

2013/1/18 JVerkuilen (Gmail) <jvverkuilen@gmail.com>:
> On Thu, Jan 17, 2013 at 1:18 PM, Mário Marques <mmarques@eeg.uminho.pt> wrote:
>> Dear All,
>>
>> I have a count data panel (xtnbreg) and I have modelled the number of
>> new foreign firms on a set of indepvars using country-pairs fixed
>> effects.
>> A referee suggested me recently to include year dummies to account for
>> time trends, the results obtained before and after the inclusion of
>> these variables are as follows:
>
> You can simply use a LR test here by fitting the nested model.
> However, you might note that there are a bunch of clusters being
> dropped for being all 0, which happens because you have no variance on
> the DV and are using a fixed effects estimator. Is this of concern?
>
>
> --
> JVVerkuilen, PhD
> jvverkuilen@gmail.com
>
> http://lesswrong.com/
>
> "Everybody loves progress but nobody likes change." ---Fortune cookie, 1/13/13.
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

--
Mário Marques
University of Minho
School of Economics and Management
Campus de Gualtar
4710-057 Braga
Portugal

Tel: +351 253 604100 (ext. 5589)
Email: mmarques@eeg.uminho.pt

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```