[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: impute question

From   Maarten buis <>
Subject   Re: st: impute question
Date   Thu, 10 Apr 2008 08:45:08 +0100 (BST)

--- joe ragg <> wrote:
> I have a multiple logistic regression model, where the predictor
> variables are binary. I wanted to impute values for missing predictor
> variables so as to make full use of the available data.  The stata
> impute command, as you all may be aware, imputes a value that is
> calculated according to a multiple regression model of the other
> predictor variables on the predictor variable of interest, and so the
> imputed values are not binary but rather between 0 and 1. I was
> wondering if this is a valid technique given my binary predictor
> variables. I would like some assurance that it is valid, as I think
> that probably it is valid. 

Unfortunately, -impute- is not valid at all. A long discussion of this
point was started here:

In short you should use -ice-, see: -findit ice-.

Royston, P. 2004. Multiple imputation of missing values. Stata Journal
4(3): 227241.

Royston, P. 2005. Multiple imputation of missing values: update. Stata
Journal 5(2): 188201.

Royston, P. 2005. Multiple imputation of missing values: Update of ice.
Stata Journal 5(4): 527-536. 

You can find these articles here:
Hope this helps,

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Yahoo! For Good helps you make a difference
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index