Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: Imputation of missing values/dealing with multiple imputations in Stata


From   "Scott Merryman" <smerryman@kc.rr.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: Imputation of missing values/dealing with multiple imputations in Stata
Date   Thu, 21 Aug 2003 21:01:11 -0500

----- Original Message ----- 
From: "Andrea Baccarelli" <andreastata5@hotmail.com>
To: <statalist@hsphsun2.harvard.edu>
Sent: Thursday, August 21, 2003 11:38 AM
Subject: st: Imputation of missing values/dealing with multiple imputations in
Stata


> I am working on a dataset where the exposure of interest was measured in the
> plasma of the study subjects. The detection limit is variable (it depends on
> the amount of plasma available for the assay)
>
> Two points:
> 1. I was suggested to use an imputation method, which assigns a value for
> each missing measurement by selecting a value from the assumed distribution
> using maximum likelihood parameter estimates, related to the approach
> described by Helsel [Helsel DR. Less than obvious - statistical treatment of
> data below the detection limit.  Environ Sci Technol 1990;
> 24(12):1766-1774.].
> I have seen that Stata has different commands to deal with missing values.
> Does anybody know whether there is any command available to perform this
> kind of imputation?
>
> 2. I need to impute missing values not only one time, but to use multiple
> imputation to reflect the uncertainty of the estimates (kind of a
> "bootstrap" procedure).
> How to perform the analysis (linear regression, or any other kind of
> analysis) on the multiple imputations? SAS 8.0 has a new procedure (PROC
> MIANALYZE, SAS 8.0) that does exactly this.
> Does anybody know whether it is possible to do it also in Stata?
>
> Thanks,
> Andrea
>
>

There are a few procedures you can use.

-whotdeck- (or -hotdeck-): Multiple Imputation using the Approximate Bayesian
Bootstrap (hotdeck) with weights

-hotdeckvar- : Creating imputed variables through single hotdeck imputation


-remsng-:  Multiple regression with missing observations for some variables

-implogit-: Logistic regression for data including multiple imputations

These files can be downloaded with -findit-


In addition, there is

AMELIA: A Program for Missing Data -- for Windows (self-extracting file;
version: 2.1, 7/15/2003) or for Gauss (zip file; version: 2.1, 7/15/2003). By
James Honaker, Anne Joseph, Gary King, Kenneth Scheve, and Naunihal Singh. Both
versions implement the same key procedures described in Gary King, James
Honaker, Anne Joseph, and Kenneth Scheve's "Analyzing Incomplete Political
Science Data: An Alternative Algorithm for Multiple Imputation" (APSR,
2001).....You may also be interested in an interactive video on Amelia, and
Clarify (see below), which will now automatically combine multiply imputed
datasets. For procedures Clarify does not handle, see Ken Scheve's MI program
(Version: 2.1).

and

CLARIFY: Software for Interpreting and Presenting Statistical Results. By
Michael Tomz, Jason Wittenberg, and Gary King; version: 2.1, 1/5/2003. This is a
set of easy-to-use Stata macros that implement the techniques described in Gary
King, Michael Tomz, and Jason Wittenberg's "Making the Most of Statistical
Analyses: Improving Interpretation and Presentation"...

available at http://gking.harvard.edu/stats.shtml


Hope this helps,
Scott


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index