[Numpy-discussion] Generating Bell Curves (was: Using normal() )

[Robert Kern]

On Fri, Apr 25, 2008 at 8:32 AM, Rich Shepard <rshepard at appl-ecosys.com> wrote:
> On Thu, 24 Apr 2008, Keith Goodman wrote:
>
>  > A Gaussian never reaches zero.
>
>  Keith,
>
>    I know, and that's why I need to find another way to draw these curves.
>  While mathematically any 'y' value < 0.2 (the default) is equivalent to
>  zero, the curves must reach zero in the figures.
>
>    Briefly, this model is used in regulatory compliance that may also end up
>  in legal proceedings. While the results are robust and not affected by the
>  curve not reaching zero, the appearance can cause problems in the regulatory
>  and legal arenas. They would be a distraction, a red herring, and consume
>  resources to explain. All this can be avoided by bell curves that reach
>  y=0.0 at the ends.

In that case, you need to search the literature of your field for
precise details on how to construct the curve that you want. As Anne
notes, "bell-shaped curve," while seemingly generic, usually specifies
Gaussians, and Gaussians do not have the properties you need. There
are any number of curves which we could (and have) suggested as
"looking bell-shaped," but if your field is as finicky as you say it
is, there must be literature giving these details. I don't think any
of us are familiar with that literature.

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless
enigma that is made terrible by our own mad attempt to interpret it as
though it had an underlying truth."
 -- Umberto Eco