Bookmark and Share

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


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

Re: st: Interpretation of margins in the presence of fixed effects


From   Scott Merryman <[email protected]>
To   [email protected]
Subject   Re: st: Interpretation of margins in the presence of fixed effects
Date   Fri, 6 Sep 2013 16:56:39 -0500

-margins- is estimating the predictive margins.  It is calculating the
average predicted value of size if all agedum were equal to 1, 2, ...
with the inum variables left as they are.

For example:

. webuse grunfeld,clear

. reg invest i.year i.com

      Source |       SS       df       MS              Number of obs =     200
-------------+------------------------------           F( 28,   171) =   29.27
       Model |  7744295.09    28  276581.968           Prob > F      =  0.0000
    Residual |  1615648.82   171  9448.23873           R-squared     =  0.8274
-------------+------------------------------           Adj R-squared =  0.7991
       Total |  9359943.92   199  47034.8941           Root MSE      =  97.202

------------------------------------------------------------------------------
      invest |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
<snip>
     company |
          2  |   -197.545   30.73799    -6.43   0.000    -258.2198   -136.8702
<snip>

------------------------------------------------------------------------------

The estimated coefficient for company == 2 is negative but the
predictive margin is positive:

. margin com

Predictive margins                                Number of obs   =        200
Model VCE    : OLS

Expression   : Linear prediction, predict()

------------------------------------------------------------------------------
             |            Delta-method
             |     Margin   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     company |
          2  |    410.475   21.73504    18.89   0.000     367.5715    453.3785
 <snip>
------------------------------------------------------------------------------

This is the same as the predicted average value of investment if all
the company observations were equal to 2:

. replace com = 2
(180 real changes made)

. predict xb
(option xb assumed; fitted values)

. sum xb

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
          xb |       200     410.475    56.20777   337.2628   540.0997

.
Scott


On Fri, Sep 6, 2013 at 8:45 AM, Dana Shills <[email protected]> wrote:
> Hi Jed:
>
> I understand the role of the constant but I am trying to figure what exactly the margins command is estimating in the presence of other dummies.
*
*   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/


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