Statalist The Stata Listserver


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

Re: st: 'Normal' condition number vs GLAMM condition number


From   "Stas Kolenikov" <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: 'Normal' condition number vs GLAMM condition number
Date   Fri, 1 Sep 2006 09:09:14 -0500

-gllamm-'s condition number is that of the information matrix /
covariance matrix of the estimated coefficients. -cmdnmb3-, it appears
to me (I've never used it), finds the condition number of the X'X
matrix, which is not particularly relevant for -gllamm- due to higher
order levels and possible non-linearities present.

See also: http://mathworld.wolfram.com/ConditionNumber.html,
http://en.wikipedia.org/wiki/Condition_number

On 9/1/06, Richard Williams <Richard.A.Williams.5@nd.edu> wrote:
>I have a question about condition numbers. When I calculate the condition
>number for my models using the add-on package cndnmb3 I get very acceptable
>values between 10 and 15. However, when I run the same models in GLAMM I get
>condition numbers between 400 and 600. Are these two different types of
>condition numbers and if so what is the acceptable range for each type and
>how do these two condition numbers differ? Or is one or the other program
>(cndnmb3 or GLAMM) calculating incorrectly the condition number?

I don't know, but if you want a tie-breaker you might try -collin-
and -coldiag2-, both of which will compute condition numbers, and, I
believe, give you different options for how to do it.  It may depend
on whether variables are being standardized first.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   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/


--
Stas Kolenikov
http://stas.kolenikov.name
*
*   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