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 at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: predict the dummy variable after ordered probit model


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: predict the dummy variable after ordered probit model
Date   Tue, 11 Dec 2012 10:00:54 +0100

On Tue, Dec 11, 2012 at 12:04 AM, Xuan Zhang wrote:
> I have a variable which takes value from 1,2,3,4,5. Like rep78 in the
> sample data. After running oprobit, I tried to estimate which category
>  each observation falls in.
>
> Certainly, it's not predict p1 p2 p3, p. It gives out the probability.
> And not  predict pscore, xb, it gives the S score.
>
> I tried estimates store, but the results only shifts between 0 and 1,
> not like 1 2 3 4 5.


Ordered models model the probabilities of ending up in each of these
categories, so that is what you can predict. You could use some
arbitrary cut-off, e.g. probability larger than 50% or the category
with the highest probability to assign each observation to one
category, but that basically implies an extra model on top of your
order probit model so I would not do that.

Hope this helps,
Maarten

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index