Spearman Rank Correlation Coefficient broken?

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Load the HAL or RI .sspace for TASA
2. run 'gs red blue SPEARMAN_RANK_CORRELATION'
3. Note that the value is not in the range [-1,1]

What is the expected output? 

for HAL:
> gs red blue SPEARMAN_RANK_CORRELATION 
5543.720074374325

> gs red blue SPEARMAN_RANK_CORRELATION
22.04261903227375

What do you see instead?
something in the range [-1,1]

Please use labels and text to provide additional information.

LSA seems to work:
> gs red blue SPEARMAN_RANK_CORRELATION
0.49931221458016195

Could the be due to the 0 values in the vectors?  These would indicate
ties, but this should cause the method to default to Pearson's correlation,
which it doesn't appear to be doing:
(for HAL)
> gs red blue PEARSON_CORRELATION 
0.9710481293972546
as this seems to work.

Original issue reported on code.google.com by David.Ju...@gmail.com on 10 Feb 2010 at 8:44

[GoogleCodeExporter]

This was due to a math precedence issue and only made itself present with large 
dimensions.  The recent merge from coals-fix addressed this problem.

Original comment by FozzietheBeat@gmail.com on 3 Aug 2010 at 4:56

• Changed state: Done