kurtosis and variance

[Francesco Potorti`]

The current version ov var has a second optional argument opt which
defaults to normalising to N-1.

However, kurtosis normalises to N, giving wrong results (see kurtosis at
 <http://www.itl.nist.gov/div898/handbook/eda/section3/eda35b.htm>
and standard deviation at
 <http://www.itl.nist.gov/div898/handbook/eda/section3/eda356.htm>)

So probably kurtosis should take an optional argument too, which
defaults to normalising to N-1.  If this is not possible, it should call
std with opt=1.

Same could apply for the third argument dim, if possible.

The same applies at least for skewness and statistics, and maybe for
others.

By the way, the @deftypefn of var does not show opt nor dim.

I can propose a changeset, if there is consensus of what should be done.

-- 
Francesco Potortì (ricercatore)        Voice: +39 050 315 3058 (op.2111)
ISTI - Area della ricerca CNR          Fax:   +39 050 315 2040
via G. Moruzzi 1, I-56124 Pisa         Email: Potorti at isti.cnr.it
(entrance 20, 1st floor, room C71)     Web:   http://fly.isti.cnr.it/