RFC: Statistics in Subsurface

Hartley Horwitz hhrwtz at gmail.com
Tue May 12 19:00:46 PDT 2020


> ---------- Forwarded message ----------
> From: Dirk Hohndel <dirk at hohndel.org>
> To: Willem Ferguson <willemferguson at zoology.up.ac.za>
> Cc: Subsurface Mailing List <subsurface at subsurface-divelog.org>
> Bcc:
> Date: Tue, 12 May 2020 14:24:28 -0700
> Subject: Re: RFC: Statistics in Subsurface
>
> On May 12, 2020, at 12:41 PM, Willem Ferguson <
> willemferguson at zoology.up.ac.za> wrote:
>
> ....snip......

> Lastly, I do not like candlestick graphs because the application in
> econometrics does not include the equivalent of a mean value. It is meant
> to indicate the limits and sometimes direction of change within a specific
> time period giving rise to the candle forming the central part of the
> graph. In my opinion a minimal box and whisker approach is more readily
> interpretable.
>
>
> I keep saying "candlestick" when I mean to say "box and whiskers". My
> mistake. You are spot on correct, the error is mine.
>

As much as I use stats in my job, I wonder what the value of knowing the
upper quartile is of my SAC rate or any other parameter that I may be
tracking.  For diving planning purposes I'd need to use my max air
consumption after filtering the dive list to get appropriate dives.   I
don't think this is providing the average user with useful data.  It's got
mathematical pureness to provide a 5-number data visualization assuming the
filtered list provides a statistically significant number of data points.
Other than that, do we care? Should we care?  I will send a sample box plot
to a diving friend of mine who's a part-time dive pro.  I'd be surprised if
he knows how to interpret it.  Histograms are more readily understood but
unfortunately take up a lot of space, and the issue of choosing the bin
size gets challenging as you've discussed when it comes to depth.

...Hartley
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.subsurface-divelog.org/pipermail/subsurface/attachments/20200512/fadf7f3c/attachment.html>


More information about the subsurface mailing list