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

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

From   René Wevers <>
Subject   RE: RE: st: Best method for imputing dichotomous variables
Date   Fri, 15 Feb 2008 15:58:24 +0100 (CET)

If both methods make sense and provide somewhat similar results, then that
is more then enough for me. The age variable is purely for control so a
reasonable confidence in the imputation, which definately follows from
what you propose here, is more then enough for me.

Again thanks for your replies and I hope not to have to bother any further.

"Met vriendelijke groeten" ;-)


Maarten buis wrote:
> --- 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
> -----------------------------------------
> 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
> now.
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

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