Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: Re: Multiple Imputation (MI)


From   Daniel Feenberg <feenberg@nber.org>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Re: Multiple Imputation (MI)
Date   Fri, 13 Jan 2012 12:42:53 -0500 (EST)



On Fri, 13 Jan 2012, john ebireri wrote:

Dear Statalist Users,

I'm sorry about the initial mix-up.

I know what secondary data is.

I'm actually asking if Multiple Imputation is suitable for handling missing observations in secondary data?

John.


You know what secondary data is, but the other list members do not. Do you mean aggregate data, such as cross-tabs? I can see a desire to impute values to the cells suppressed in Census publications for confidentiality reasons.

Sometimes "secondary data analysis" means the analysis is secondary, but the data is the same, in which case MI would be applicable as it was to the primary data.

But when is it applicable at all? Is there some place to find a defense of MI? I have never been able to figure out how to tell if any particular use of MI created data was valid or misleading.

Daniel Feenberg


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index