Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Trang Nguyen <nqtrang.hanoi@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: multiple imputation of complex survey data |
Date | Sat, 3 Aug 2013 12:17:47 -0400 |
Thanks much, Stas, for clarifying this, and for the article you suggested for proper analysis of imputed survey data. Regarding the imputation stage, I guess this means I need to figure out ways to manually build information about the survey's design into the imputation model. I appreciate any advice about how to do this or if there are best/recognized approaches that have been documented. Thanks much! Trang On Fri, Aug 2, 2013 at 9:39 PM, Stas Kolenikov <skolenik@gmail.com> wrote: > This is only for the analysis stage. For proper analysis of survey > data with imputations, see Shao & Sitter (1996, > http://www.citeulike.org/user/ctacmo/article/1269394). > > -- Stas Kolenikov, PhD, PStat (ASA, SSC) > -- Senior Survey Statistician, Abt SRBI > -- Opinions stated in this email are mine only, and do not reflect the > position of my employer > -- http://stas.kolenikov.name > > > > On Fri, Aug 2, 2013 at 4:02 PM, Trang Nguyen <nqtrang.hanoi@gmail.com> wrote: >> Hi! >> >> I am using complex survey data and need to do multiple imputation on >> missing values. In Stata 12 manual, there is the option to declare >> data as mi data and survey data using >> >> mi set mlong >> mi svyset ... >> >> If I do this and then do mi impute, will the imputation incorporate >> the fact that the data have survey design features (strata, cluster >> and weights)? >> >> Or is the purpose of mi svyset only to allow analysis of multiply >> imputed data as survey data? >> >> Thanks for your help clarifying this. The manual does not make this clear. >> >> Best wishes, >> >> Trang >> * >> * 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/ * * 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/