RE: st: regression with categorical predictors

["Feiveson, Alan H. (JSC-SK311)" <alan.h.feiveson@nasa.gov>]

Amir - You can tell that there is no adjustment because the reported P-values just come from the t- (or sometimes normal) distribution. In your previous example for region 3, note that the P-value of 0.538 is just P(T > -0.62) where T ~ t(45 df)


. local tval = -1.438452/2.320244

. di `tval', tprob(45,`tval')
-.61995721 .53841281

Al Feiveson


      Source |       SS       df       MS              Number of obs =      50
-------------+------------------------------           F(  4,    45) =   37.21
       Model |   4096.6093     4  1024.15232           Prob > F      =  0.0000
    Residual |  1238.40987    45  27.5202192           R-squared     =  0.7679
-------------+------------------------------           Adj R-squared =  0.7472
       Total |  5335.01916    49  108.877942           Root MSE      =   5.246

------------------------------------------------------------------------------
       drate |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      medage |   4.283183   .5393329     7.94   0.000     3.196911    5.369455
             |
      region |
          2  |   .3138738   2.456431     0.13   0.899    -4.633632     5.26138
          3  |  -1.438452   2.320244    -0.62   0.538    -6.111663    3.234758
          4  |  -10.90629   2.681349    -4.07   0.000    -16.30681   -5.505777
             |
       _cons |  -39.14727   17.23613    -2.27   0.028    -73.86262   -4.431915
------------------------------------------------------------------------------






-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of amir gahremanpour
Sent: Tuesday, August 30, 2011 2:30 PM
To: statalist
Subject: RE: st: regression with categorical predictors

thank you Richard,

I am aware of "test" and "mtest" option, I wanted to make sure I need to use it (not to adjust twice) !

Amir


----------------------------------------
> Date: Tue, 30 Aug 2011 15:23:09 -0500
> To: statalist@hsphsun2.harvard.edu
> From: richardwilliams.ndu@gmail.com
> Subject: Re: st: regression with categorical predictors
>
> At 11:06 AM 8/30/2011, amir gahremanpour wrote:
> >Hello everybody,
> >
> >First let me apologize if my question is very primitive, I am trying
> >to self educate myself about regression with stata.
> >When I have a categorical predictor with more than two levels , I am
> >not sure if the p-values stata provides are adjusted for multiple
> >comparison or are not? This example is from stata documentation ,
> >variable region has 4 levels , are the p-values adjusted for 3
> >comparisons to the base level? I do not think so !
>
> You may wish to test whether the dummies as a group are significant,
> e.g. in your example you could use either of these commands:
>
> testparm i(2/4).region
> test 2.region 3.region 4.region
>
> You might also do
>
> test 2.region 3.region 4.region, mtest(b)
>
> See the help for test.
>
>
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> HOME: (574)289-5227
> EMAIL: Richard.A.Williams.5@ND.Edu
> WWW: http://www.nd.edu/~rwilliam
>
> *
> * 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/

*
*   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/