Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: how to fill some gross wage missings if I have the net wage


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: how to fill some gross wage missings if I have the net wage
Date   Thu, 16 Dec 2010 10:54:56 +0000 (GMT)

--- On Thu, 16/12/10, Laura R. wrote:
> I came up with the following solution: just make a
> regression on gross wage with several independent
> variables with the cases that report gross wage, and
> then calculate the gross wage for the others using the
> b coefficients and the independent variables. Would that
> work as well?

No, if you want to do imputation you'll have to do it 
right, i.e. do multiple imputation. What you propose would
put all imputed observations on the regression line, i.e.
you would overestimate the correlations and all subsequent
statistics that depend on them. Moreover you are treating
your imputations as real observations rather than results
from a model. To do this right you must also take the 
imputation model uncertainty into account. 

In Stata 11.1 you can use panel data (-xt-) models with 
-mi estimate-, so the presence of panel data is not a 
reason to forgo multiple imputation.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

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



      

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index