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

RE: RE: st: Best method for imputing dichotomous variables

From   Maarten buis <>
Subject   RE: RE: st: Best method for imputing dichotomous variables
Date   Fri, 15 Feb 2008 14:35:21 +0000 (GMT)

--- René Wevers <> wrote:
> I also want to impute a discrete variable, namely the age of
> companies in years (integers) with a maximum of 37 years (age has
> only been measured as of 1967). The distribution is for this variable
> is definately not normal, but it is not extremely skewed as well.
> Can I manipulate my data such that I can apply standard (OLS)
> regression to impute with -ice- or should I apply ordinal logistic
> regression in this case?

You could have a look at -gladder- which compares various
transformations against a normal curve. For normality checking I prefer
-hangroot-, downloadable from -scc-, over a histogram with a normal
density plotted through it. (though I am not completely neutral as I am
the author of -hangroot-.) I have also heard request for StataCorp to
increase the number of categories allowed in -ologit- for just the kind
of imputation analysis you are proposing, so there are other persons
who think that it is reasonable to use -ologit- this way. I would say
try to transform age into something more nearly Gaussian and use
-regress- and use -ologit- and see if it makes a difference.

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! Answers - Got a question? Someone out there knows the answer. Try it
*   For searches and help try:

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