RE: st: Grouping income variables- RECODE COMMAND

["Antonio Rodriguez Andres" <Antonio.Andres@emu.edu.tr>]

Maarten

Thank you for detailed response. One of the reviewers was concerned about
the missing values in our income
Variable. The variable dprt is the depression score across individuals
taking values from 0 to 24. How can I observe that the missingnes
Is independent of the depression score?
Antonio

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Maarten Buis
Sent: Tuesday, February 04, 2014 4:09 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Grouping income variables- RECODE COMMAND

On Tue, Feb 4, 2014 at 2:54 PM, Antonio Rodriguez Andres wrote:
> What you are suggesting is to use the command mi?
> which variables should be included) all the variables included in the
original regression?

You don't have to do anything as long as you can reasonably believe that the
probability of missingnes is independent of dprt (whatever that may be).
Notice that this is something you can see in your data.
In many situations, that is the safest solution. If you need a reference,
you can cite: Allison, Paul D.  (2001). Missing data.
Thousand Oaks: Sage.

Multiple imputation would be my second choice. You would have to include all
variables in your model (including your dependent
variable!!) in imputation model.

But all this depends on the details of the problem.

-- Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

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


*
*   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/