matrix rank

[Brian Gough]

Robert G. Brown writes:
 > I'm getting shot down by antispam software every time I post, so I'll
 > make this up out of several short messages.  First, for list
 > statisticians -- I need to evaluate the rank of an MxN (possibly square)
 > integer matrix.  Is the best routine for this purpose one of the SVD
 > routines or something else?  I could write one, but I'm trying to use
 > GSL code where I can as it is likely better than a simple pivoted
 > elimination (and I don't have to write it:-)

For practical work the SVD with a suitable cutoff is likely the best
choice.

-- 
Brian Gough

Network Theory Ltd,
Commercial support for GSL --- http://www.network-theory.co.uk/gsl/