Antwort: st: margins after logit model

[Justina Fischer <JFischer@diw.de>]

Hi

you have just three years.

A very easy way is to code two dummies for the years 2005 and 2007 each - the year not coded (2006) is then your base year in your regression.

Cheers

Justina



-----owner-statalist@hsphsun2.harvard.edu schrieb: -----

An: <statalist@hsphsun2.harvard.edu>
Von: Haiyong Xu <hxu@mednet.ucla.edu>
Gesendet von: owner-statalist@hsphsun2.harvard.edu
Datum: 30.12.2010 12:36AM
Thema: st: margins after logit model

Hi,
Wish you all have wonderful holidays!

My logit model and the output are as follows:

logit psydrgnobzn i.female age i.year, vce(cluster cluster_id)

Logistic regression                               Number of obs   =
40284
Wald chi2(4)    =
422.81
Prob > chi2     =
0.0000
Log pseudolikelihood = -26350.649                 Pseudo R2       =
0.0127

(Std. Err. adjusted for 13428 clusters in
cluster_id)
------------------------------------------------------------------------------
|               Robust
psydrgnobzn |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
1.female |   .5114119   .0357798    14.29   0.000     .4412848
.5815389
age |  -.0191765   .0023834    -8.05
0.000    -.0238479   -.0145051

|
year |

2006  |    .154101   .0147308    10.46   0.000     .1252292
.1829728
2007  |   .2495566   .0171274    14.57   0.000     .2159875
.2831257

|
_cons |   .5000495   .1866511     2.68   0.007       .13422

.865879
------------------------------------------------------------------------------

After the logit model, I run margins command to get the marginal effect.
. margins, dydx(year)

Average marginal effects                          Number of obs   =
40284
Model VCE    : Robust

_expression_   : Pr(psydrgnobzn), predict()
dy/dx w.r.t. : 2006.year 2007.year

------------------------------------------------------------------------------
|            Delta-method
|      dy/dx   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------

year |
2006  |   .0351202   .0033485    10.49   0.000     .0285573

.0416832
2007  |   .0575292   .0039259    14.65   0.000     .0498345
.0652239
------------------------------------------------------------------------------
Note: dy/dx for factor levels is the discrete change from the base level.

The year variables has three values 2005, 2006 and 2007. Stata automatically
set 2005 as the omit group. I wonder if I can set 2006 as the reference
group so that I can get the margin effect between 2006 and 2007.

Thanks.
Haiyong



IMPORTANT WARNING:  This email (and any attachments) is only intended for the use of the person or entity to which it is addressed, and may contain information that is privileged and confidential.  You, the recipient, are obligated to maintain it in a safe, secure and confidential manner.  Unauthorized redisclosure or failure to maintain confidentiality may subject you to federal and state penalties. If you are not the intended recipient, please immediately notify us by return email, and delete this message from your computer.

*
*   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/