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/