Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: post-regression diagnostics for LPM


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: post-regression diagnostics for LPM
Date   Fri, 8 Apr 2011 14:51:26 +0100

I guess that you are just using -regress-. You don't give any details
of commands used. Either way, whatever your model command, you can
probably just go

predict myprob if e(sample)
su myprob, detail
count if !inrange(myprob,0,1)

But you should check whether what you want is the default output of -predict-.

I wouldn't stop there. Two easy diagnostic plots are

rvfplot

and

ovfplot

The last requires installation of -modeldiag- from SSC.

The question is not just whether the model yields absurd answers for
your estimation sample, but also whether it would do so for plausible
values that you did not observe. That is trickier.

The Statalist FAQ requests use of real names in communicating with the list.

Nick

On Fri, Apr 8, 2011 at 2:40 PM, traumaprof <traumaprof@gmail.com> wrote:

> I need to compute the percent of predicted probabilities that fall
> outside the 0-1 range in order to tell whether LPM will work for a
> data set. Any recommended commands? Thanks
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index