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

Re: st: RE: linear probab model


From   "Leopold Remi Sarr" <lrs25@cornell.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: RE: linear probab model
Date   Thu, 18 Sep 2003 14:09:41 -0400 (EDT)

Dear Nick,

I am aware of the fact that the lpm may lie outsie the [0,1] range but I
would like to be able verify that the estimated lpm is appropriate for the
data I am using. I know that there is standard way of checking it with a
logit model by setting a set of 2 equations and testing the original model
with those additional regressors and running the test on those 2
variables. However, I have no clue how to do it with the lpm or the
probit.



Leopold Remi Sarr
>
>> Does anyone know how to test that a linear probability
>> model correctly
>> represents the data. Idem for a probit. Thanks for your help,
>
> I'd like to know what a correct representation
> of the data is. In one sense, only the data
> preserve the data.
>
> Nevertheless here is one suggestion:
>
> if the predictions from a linear probability
> model go outside [0,1] within the observed
> range of the data, think again.
>
> However, as this is the standard objection
> to said model rehearsed in every text,
> I suspect you seek a much deeper answer.
>
> Nick
> n.j.cox@durham.ac.uk
> *
> *   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 1996–2022 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index