Statalist


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

Re: st: AW: Help with ologit


From   Richard Williams <Richard.A.Williams.5@ND.edu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>, Stata <statalist@hsphsun2.harvard.edu>
Subject   Re: st: AW: Help with ologit
Date   Mon, 14 Dec 2009 11:27:58 -0500

At 11:06 AM 12/14/2009, Daniel J. Quinones wrote:
I am using Stata 10.1.
At first I was simply using "mfx compute," as I had done with a probit
model.  I am now using "mfx compute, predict outcome(#))" where # represents
an outcome of my dependent variable.  The latter method provides results.

Again, -margeff- or -mfx2- can simplify your life, by automating the process so you get every outcome with a single command. -margeff- is also faster than -mfx- and offers additional powerful options.

Is there a function for ologit similar to what lstat does for probit
(specifically to get a % for correctly predicted)?  I would like to get some
idea of my model's reliability.

Not to my knowledge. I've never been that fond of -lstat- anyway. Especially if the outcome is relatively rare, or common, you can wind up having all the cases be predicted as a 1 or else as a 0.

You might consider the -leastlikely- command, which is part of Long and Freese's -spost9- package. It will help to identify those cases where the model seems to be having the most trouble.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   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   |   What's new   |   Site index