Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: out-of-sample predictions with GLM


From   jc1926@gmx.de
To   statalist@hsphsun2.harvard.edu
Subject   st: out-of-sample predictions with GLM
Date   Thu, 26 Jul 2007 14:00:18 +0200

Greetings,

Using proportions data, I am estimating a GLM of the type described in Papke and Wooldridge's 1996 article entitled "Econometric methods for fractional response variables...", Journal of Applied Econometrics, vol 11, 619-632.

The Stata command to estimate the model is:
glm y x1 x2, fam(bin) link(logit) robust

One nice feature of the model is that the in-sample predictions (logically) sum to one. I would like to use the model to calculate out-of-sample predictions. In this case, the predicted shares do not sum to one, but are sometimes higher or lower depending on the values for x that I plug in. This is also logical, but I'm just wondering how I interpret the predicted shares.

Is it OK to simply weight them accordingly so that they sum to one? Also, can any intepretation be given to the sum of the predicitons (e.g. if greater than 1, can it be said that underlying stock is predicted to increase)? 

Many thanks for any insights,

JC
-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index