Re: st: post estimation and standard errors

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Fri, 11/6/10, Stefan wrote:
> Maybe someone can help. I want to meddle with my standard
> errors after regress to see how this affects the Wald test.
> Since Stata does not let me simply change the system
> variables _se[VAR] and then run the test and the Wald test
> takes only
> standard adjustments for the SEs as an argument I see no
> option besides coding or doing it manually. Is there a nicer
> way?

the graphs below show the consequences of changes in the 
standard error for the Wald statistic and resulting the 
p-value, assuming that the null hypothesis is that the
coefficient equals 0.

*------------------ begin example ----------------
sysuse auto, clear
logit foreign mpg rep78
twoway function y = _b[mpg]/x, ///
       range(.05 .5)           ///
       ytitle(Wald statistic)  ///
       xtitle(standard error)  ///
       xline(`=_se[mpg]')


twoway function y = 2*normal(-abs(_b[mpg]/x)), ///
       range(.05 .5)                           ///
       ytitle(p-value)                         ///
       xtitle(standard error)                  ///
       xline(`=_se[mpg]')
*------------------ end example --------------------
(For more on examples I sent to the Statalist see: 
http://www.maartenbuis.nl/example_faq )

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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/