Re: New Statistics functions

[Nir Krakauer]

Thanks for the feedback, Arno!

> 1) Could you use corr instead of corrcoef to make the function more
> future-proof?
>
> 2) If name/value pairs were used as arguments for penter and premove,
> then full Matlab compatibility could be added later without breaking any
> code that might be written for the current argument structure.

I used the syntax I'm more comfortable with, but if you or someone
else wants to make these changes, I certainly won't object.

Nir