Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: intraclass correlation for Categorical data


From   John Litfiba <cariboupad@gmx.fr>
To   Dirk Enzmann <dirk.enzmann@uni-hamburg.de>
Subject   Re: st: intraclass correlation for Categorical data
Date   Fri, 13 May 2011 00:49:27 +0200

Many thanks Dirk,

But this doesnt work, as I have more than 2 categories ...
To be more precise, I have a panel dataset which is too large for
using random or fixed LOGIT models...
Therefore I am constrained to fit a multinomial logit (which works
fine even with millions of observations), and I was trying to figure
out if the "correlation" between the categorical observations within
my clusters (the individuals of my panel dataset) show a high or low
level of correlation... I hope of course to find a low "correlation"
(can we say "correlation" for nominal variables? not sure) in order to
justify the independence assumption of the multinomial logit ;-)
Maybe there is no short answer to this problem... ;)

Best,

On 13 May 2011 00:08, Dirk Enzmann <dirk.enzmann@uni-hamburg.de> wrote:
> With respect to dichotomous data, see -xtmrho- on ssc.
>
> Dirk
>
*
*   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   |   Site index