Bookmark and Share

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

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

R: st: control a variable in stata

From   <>
To   <>
Subject   R: st: control a variable in stata
Date   Sun, 22 Apr 2012 14:20:51 +0200

Andy may want to check for heteroskedasticity after -regress- via - estat
hettest- ( from -regress postestimation- suite).
As an aside, Huber-White sandwich estimator is implemented via the -robust-
option available with most Stata commands (including -regress-), as reported
in Baum CF. An Introduction to Modern Econometrics Using Stata. College
Station, TX: Stata Press, 2006: 136-38.
Best wishes,

-----Messaggio originale-----
[] Per conto di Kong, Chun
Inviato: domenica 22 aprile 2012 01:05
Oggetto: RE: st: control a variable in stata

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

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


[] on behalf of Nora Reich
Sent: 21 April 2012 20:38
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.


Am 21. April 2012 21:33 schrieb Nora Reich <>:
> 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-
> (for more information type -help histogram-).
> 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
> Publications:
> ml

Nora Reich

*   For searches and help try:

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index