RE: st: stat query

[Nick Cox <n.j.cox@durham.ac.uk>]

There was a Stata-based review in 

SJ-8-3  pr0041  .  Speaking Stata: Corr. with confidence, Fisher's z revisited
        (help corrci, corrcii if installed) . . . . . . . . . . . .  N. J. Cox
        Q3/08   SJ 8(3):413--439
        reviews Fisher's z transformation and its inverse, the
        hyperbolic tangent, and reviews their use in inference
        with correlations

Perhaps the most useful detail for Rich was the aside that atanh[r] = (1/2) logit[(1 + r)/2]. 

Hence scaling the correlations to fall inside [0,1] and logit modelling the result opens up many routes in Stata. This is evident otherwise, but it is reassuring nevertheless that the nice properties of Fisher's z are akin to those to logit. 

Nick 
n.j.cox@durham.ac.uk 

Richard Goldstein

thank you very much (not sure how I forgot about this, but ...)

Marcello Pagano 

> http://en.wikipedia.org/wiki/Fisher_transformation
 
On 11/18/2010 2:21 PM, Richard Goldstein wrote:

>> a client has told me that he thinks he will soon have a project where
>> the dependent (response) variable will be a correlation coefficient;
>> i.e., a "continuous" variable with the range restricted to [-1,+1];
>>
>> any help, guidance, suggestions on what model might be appropriate for
>> this would be greatly appreciated; similarly, any citations would be
>> great appreciated

*
*   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/