Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: out-of-sample prediction tables


From   Eleimon Gonis <Eleimon2.Gonis@uwe.ac.uk>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: out-of-sample prediction tables
Date   Thu, 19 Jan 2012 16:29:03 +0000

Dear Statalisters,

I'm running Stata/IC 11.2 on Windows XP. 

I have data for 10 years (assume year variable: year from 1 to 10) and I want to fit a simple ologit/oprobit model (ologit Y X).

I then want to repeat the exercise but for two sub-periods, i.e. year<6 and then year>=6. Then, I want to use each fitted model to calculate the expected percentage correctly predicted for each year out of the sample, e.g. use the (year<6) fitted model to predict classification for years 6, 7, 8, 9 and 10 respectively. In other words, I want to build classification tables, much like the "estat classification" command does for binary logits/probits or the "epcp" routine by Christopher Lawrence. Nevertheless, the estat clas command only applies to binary models, if I'm right, and the epcp routine does not allow for the if option to be added.

Any ideas as to how I can achieve that?

Also, would anything change if I want to run the same as a panel ologit/oprobit?

Thank you,
Eleimon



*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index