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


From   Uri Goldbourt <[email protected]>
To   [email protected]
Subject   st: RE: LOGIT MODEL
Date   Mon, 31 Dec 2007 17:56:51 +0200

Look in the brand new issue of STATISTICS IN MEDICINE. (Published by Wiley).

This is the main theme of the entire issue.





-----Original Message-----
From: owner-[email protected] [mailto:owner-[email protected]] On Behalf Of Ismail Ait Saadi
Sent: Monday, December 31, 2007 8:55 AM
To: [email protected]
Subject: st: LOGIT MODEL


Dear All,


I know most of you are on holiday but I really appreciate your help and time.


I am running a logit model to predict currency crises. I have 14 variables. I run MODEL1 with all 14 variables, then run MODEL2 with only 7 variables which are mostly used by researchers. I wanted to compare the two models and see which one is superior.


1-Can you please give me the list of indicators that can be used for such comparison?


2-I know that ROC and PSEUDO R2 can be used but I get confused when some indicators are in favor of MODEL1 and others are in favor of MODEL2. What should I do.


3- What are the general rules in comparing two logit models.



Many thanks.

Enjoy your Holidays,





*   For searches and help try:




© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index