st: R: Missing data in dataset, long format

["Carlo Lazzaro" <carlo.lazzaro@tiscalinet.it>]

I I would say that the core question with missing data is indeed the one Christina seemingly prefers to side-track (is missingness in her data ignorable or not?).
That said, Christina may want to consider hot-deck imputation, that can be briefly described as replacing "each missing value with a random draw from a subsample of respondents that scored similarly on a set of matching variables." (Source: Enders CK. Applied Missing Data Analysis. New York: Guilford, 2010: 49).
However, the same source warns about biased correlations and regression coefficients and possible underestimates of standard errors. Jacknife standard error was proposed for working around this last drawback that hot deck imputation might bring about.

Best regards,
Carlo
-----Messaggio originale-----
Da: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Christina Wei
Inviato: giovedì 21 novembre 2013 19:48
A: statalist@hsphsun2.harvard.edu
Oggetto: st: Missing data in dataset, long format

I have a rather odd question for the group, but currently I am dealing with a dataset that has only a few missing data.  Without going into the details of exploring patterns of "missingness," is there a simple method for filling in missing data without engaging in any fancy methods like multiple imputation?  My goal is to crudely study my database right now before I embark on intense statistical analysis.

Also, my data is in long format, if that makes any difference.

Thanks,
Christina
*
*   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/