Recommendation for Octave statistics functions

[Ben Abbott]

On Jan 25, 2008, at 1:51 AM, John W. Eaton wrote:

> On 24-Jan-2008, Ben Abbott wrote:
>
> | How will the determination be made as to what are core functions and
> | non-core functions?
>
> The definition is a bit loose, but if a function is a core function in
> Matlab, it should probably be a core function in Octave.  People
> porting existing code will probably expect that.  I'm willing to be
> flexible on this, but it seems clear to me that specialized sets of
> functions like the ones in the quaternion, finance, and control
> directories belong in separate packages.  Where the optimization,
> statistics, and signal functions belong is not as obvious (to me,
> anyway).  For statistics, the basic things like std, mean, var,
> median, etc. should probably be included.  The more advanced things,
> maybe not.
>
> jwe

My thoughts are similar. The area that is cloudy for me are those  
Mathworks functions that are included in their toolboxes, but are not  
core to Matlab. Very often many of these functions have wide appeal  
and value.

There are some very good examples in the Mathworks optimization  
toolbox. However, there are also some good examples in the statistics  
toolbox ... quantile for example ... ok, I'm shamelessly plugging for  
my recent work, but I do think it a good example.

Thoughts?

Ben