[R] Spearman correlation test confidence intervals

[Peter Dalgaard]

Mahesh Krishnan <mskrishnan71 at yahoo.com> writes:

> Hello,
> 
> How does one calculate the confidence levels for a spearman correlation
> test?  I noticed cor.test(x,y) gives the confidence levels only for
> pearson test.
> 
> Thank you for the suggestions, in advance.

You can't. Not easily anyway. The distribution of the Spearman
correlation is known under the null hypothesis of independence
(through symmetry considerations), but not otherwise. In largish
samples, I suppose that resampling techniques could be employed.

-- 
   O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark          Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)                  FAX: (+45) 35327907