Re: st: RE: Correlation ratio based on entropy

[Kristian Bernt Karlson <kristian.karlson@gmail.com>]

Nick,
Thanks. I got at the same result. Luckily I already did the 
calculations manually, but I wanted to double-check that I got them 
right.
Kristian

Den 10. oktober 2011 13:01:53, Nick Cox skrev:
> -findit- results imply that you need to program it yourself.
>
> Nick
> n.j.cox@durham.ac.uk
>
> Kristian Bernt Karlson
>
> I have been searching on Statalist for any ado-files estimating the
> correlation ratio based on entropy (an "R-squared" type of statistic) as
> described by Magdison (Social Science Research 1981) and Haberman (JASA
> 1982), and also mentioned by Agresti (Technometrics 1986). This is a
> correlation ratio for nominal outcomes. Does anyone know whether I can
> obtain this ratio from a Stata command?
>
>
> *
> *   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/
*
*   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/