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

st: RE: Re: Predicted value and dummy explantory variables


From   "Maarten Buis" <M.Buis@fsw.vu.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Re: Predicted value and dummy explantory variables
Date   Tue, 15 Mar 2005 16:31:58 +0100

Dawat,

I would not worry too much about using the mean of a dummy variable, especially if you just want to control for gender differences and are not interested in describing them. An alternative that will offend (even) less people is to choose the modal category for dummy variables. 

An issue that requires more attention in cases like these is choosing the right values for polynomial and interaction terms, use 
mean(x)^2 instead of mean(x^2) and mean(x1)*mean(x2) instead of mean(x1*x2). 

Maarten

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of Friedrich Huebler
Sent: dinsdag 15 maart 2005 16:06
To: statalist@hsphsun2.harvard.edu
Subject: st: Re: Predicted value and dummy explantory variables

Clive,

-mfx- calculates the marginal effects at the mean values of the
independent variables. Evaluating a male-female dummy variable at its
mean is therefore not as unusual as you think.

Friedrich Huebler

--- Clive Nicholas <Clive.Nicholas@newcastle.ac.uk> wrote:
> Dawat Lidia wrote:
>
> > I am estimating OLS regression model. From this I was trying to
> > estimate the
> > predicted value holding regressors at their mean value. But I am
> > not clear
> > if I can also hold dummy regressors at their means value. Is it
> > justifiable
> > to put dummy explantory variables at their mean value and obtain
> > the predicted value ?
>
> No, it isn't: how can it be? If your dummy variable is gender, how
> can it
> possibly make sense to estimate the effect of being 0.47 of a male
> on whatever the response of interest is?
>
> A quick solution to your problem is to pop along to -whelp adjust-.
> There
> you'll find that you can hold your continuous predictor and
> covariate
> variables at their means (if you desire), whilst setting your
> dummies at 0 or 1 (or both, if you choose to -adjust- twice).
>
> CLIVE NICHOLAS        |t: 0(044)7903 397793
> Politics              |e: clive.nicholas@ncl.ac.uk
> Newcastle University  |http://www.ncl.ac.uk/geps



               
__________________________________
Do you Yahoo!?
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/
*
*   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/

*
*   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