st: re: management of missing

[Christopher Baum <baum@bc.edu>]

<>
Rodrigo said

My question to the list is if Stata allows you to avoid dropping  
observations based on the listwise criteria. I want to know this  
because maybe this will be a possible solution for raising the sample  
where the estimations are based.

No, for obvious reasons. Consider linear regression, where you  
calculate X'X. Consider an X matrix, N x k, with some random elements  
of each column missing. What does it mean to calculate sums of squares  
and cross products of such a matrix? Calculating those sums over only  
the feasible (pairwise) terms will not give you anything sensible  
except in the limit, and even asymptotically, if the proportion of  
missing data is fixed, driving N -> \infty probably won't help.

It sounds like you might be a good candidate for Stata's new manual on  
Multiple Imputation.

Kit

Kit Baum   |   Boston College Economics and DIW Berlin   |   http://ideas.repec.org/e/pba1.html
An Introduction to Stata Programming   |   http://www.stata-press.com/books/isp.html
An Introduction to Modern Econometrics Using Stata   |   http://www.stata-press.com/books/imeus.html

*
*   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/