# Re: st: RE: first stage prediction gets negative value for positive observations

 From Mark Schaffer To statalist@hsphsun2.harvard.edu, yi jiang Subject Re: st: RE: first stage prediction gets negative value for positive observations Date Mon, 21 Feb 2005 23:35:11 +0000 (GMT)

```This can happen when you use a linear model for something that is nonlinear
(in this case, that has inequality constraints).  The best known example, in
my field at least, is the linear probability model.  Have a look at any
econometrics textbook that discusses the LPM, and you'll see that the
drawback of predicted probabilities that are <0 or >1 is mentioned.  This is
one reason why logits and probits are usually preferred to the LPM.

Hope this helps.

--Mark

Quoting yi jiang <yistata@gmail.com>:

> thank you Nick. even though i plot the data and the fitted data,
> the
> problem is still there. let me give some examples:
> 1. X1>0, however, the fitted value of X1 from first stage has
> negative
> observation.
> 2. 0<X1<1, however, the fitted value of X1 from first stage has
> observation bigger than 1.
> you name it.
> I guess it doesn't matter, but it does look questionable in terms
> of
> how to explaine the meaning of X1.
>
> On Mon, 21 Feb 2005 22:47:39 -0000, Nick Cox <n.j.cox@durham.ac.uk>
> wrote:
> > Why not plot the data and your fit,
> > if that makes sense?
> >
> > Nick
> > n.j.cox@durham.ac.uk
> >
> > yi jiang
> >
> > > I'm doing a two stage regression. I'm predicting X1 (X1 are
> all
> > > positive) at the first stage and then use the predicted value of
> X1 at
> > > the second stage.
> > > However, the predicted value of X1 shows some negative value,
> which
> > > makes no sense in terms of data--X1 cannot be negative. any
> thoughts?
> >
> > *
> > *   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/
>

Prof. Mark Schaffer
Director, CERT
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS
tel +44-131-451-3494 / fax +44-131-451-3294
email: m.e.schaffer@hw.ac.uk
web: http://www.sml.hw.ac.uk/ecomes
________________________________________________________________

DISCLAIMER:

This e-mail and any files transmitted with it are confidential
and intended solely for the use of the individual or entity to
whom it is addressed.  If you are not the intended recipient
you are prohibited from using any of the information contained
in this e-mail.  In such a case, please destroy all copies in
Watt University does not accept liability or responsibility
for changes made to this e-mail after it was sent, or for
viruses transmitted through this e-mail.  Opinions, comments,
conclusions and other information in this e-mail that do not
relate to the official business of Heriot Watt University are
not endorsed by it.
________________________________________________________________
*
*   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/
```