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

# RE: st: control a variable in stata

 From "Kong, Chun" To "statalist@hsphsun2.harvard.edu" Subject RE: st: control a variable in stata Date Sat, 21 Apr 2012 23:04:31 +0000

```Thank you very much for all your help!

I have run both poisson and OLS, the OLS gives a R2 of 0.6175 and Poisson has a R2 of 0.6250, however, all the research that I have gone through is using OLS. Therefore, I think I should go with OLS, but I really appreciate for your suggestion and time.

Most of the reserachs stressed in desribing the variable and analyzing the resutls, however, very few have explained the methodology. One have used white standard corrected errors because there is a difference between the adjusted standard errors and the normal standard errors, suggesting there is small level of heteroskedascticity. Therefore, the model is regressed using the white standard errors. I have google something relate to white standard error, however, i still do not understand whether i should follow this approach.

I am sorry for all the silly question. Once again, thank you very much for your time and help. :)

Andy

________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Nora Reich [nhmreich@googlemail.com]
Sent: 21 April 2012 20:38
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: control a variable in stata

Andy, by the way, what is the value of the R2? It gives you a hint on
how much of the variation in salary can be explained by your model. I
would improve the model with new variables (as proposed earlier) and
see which ones have a significant effect and which ones lead to
substantial increases in R2.

Nora

Am 21. April 2012 21:33 schrieb Nora Reich <nhmreich@googlemail.com>:
> As far as I know, -poisson- is for skewed distributions, and salary
> distribution in general often fulfills this requirement, but salary of
> NBA players might show a different distribution. The distribution can
> be checked, e.g. with the command
>
> -histogram salary-
>
>
> Andy, I would compare the assumptions and requirements of different
> estimation strategies (OLS, poisson) and find out which fits better
> with the data.
>
> I would also check which estimations strategies are used by similar
> papers, and why.
>
> Apart from -poisson- and -regress-, I cannot think of any at the
> moment that have to be considered for your salary-estimations.
> However, if there was something like a "minimum wage" for NBA players,
> i.e. salary is censored,  -tobit- would be an alternative.
>
> Best regards
> Nora
>
>
>
>
> --
> Nora Reich
> www.nora-reich.de
> Publications:
> http://www.nora-reich.de/publikationen.html
> http://www.hwwi.org/ueber-uns/team/forscher/nora-reich/publications.html

--
Nora Reich
www.nora-reich.de
Publications:
http://www.nora-reich.de/publikationen.html
http://www.hwwi.org/ueber-uns/team/forscher/nora-reich/publications.html

*
*   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/

*
*   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/
```