[ncl-talk] bug report stddev (build-in function)

[Mary Haley]

Hi Leo,

I just checked the stddev documentation, and didn't see the same mention of
"population" that you did.

The first line of this page states:

*     "Calculates the sample standard deviation."*

It does mention population later in this documentation, and I need to look
into why this is. I think this is an error in the documentation.

As for the "stdMonTLL" function, it uses dim_stddev under the hood, so this
should also be sample standard deviation.

--Mar

On Wed, Jan 28, 2015 at 3:00 AM, Kampenhout, L. van (Leo) <
L.vanKampenhout at uu.nl> wrote:

>
>   Hi,
>
> I just stumbled upon the following inconsistency. In the description of
> function stddev it is written that (
> http://www.ncl.ucar.edu/Document/Functions/Built-in/stddev.shtml)
>
> *  (…) this function calculates the population standard deviation.*
>
> However, I’m afraid it does not compute the *population* std but rather
> the *sample* std. Example program attached.
>
> I see two solutions:
> (1) either change the documentation, such that is states specifically this
> function computes the sample standard deviation (perhaps also add an
> explanation why this is better).
> (2) change the code . This will create a backward incompatibility and many
> user scripts may yield different results.
>
>  As a side note, function stdMonTLL also computes the sample std, second
> example program attached.
>
> Kind regards,
> Leo
>
>
>
> ____________________________________________________________
>
> Leo van Kampenhout
> Institute for Marine and Atmospheric Research (IMAU), Utrecht University
> BBG room 672
> Princetonplein 5, 3584 CC Utrecht, The Netherlands
> +31 30 2533271
> L.vanKampenhout at uu.nl
> www.uu.nl/wce
>
>
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150128/80f89a43/attachment.html