[PLUG-TALK] Excel Formula

[Rich Shepard]

On Mon, 22 Jan 2007, M. Edward (Ed) Borasky wrote:

> Yeah ... I knew there were bugs in Lotus 1-2-3, which I always assumed
> were "upward compatible" from Visicalc. My favorite is the day number
> convention, which assumed that 1900 was a leap year -- which it wasn't. :)

   Now that you mention it, I remember that, too.

> Excel at least has both sample standard deviations and population standard
> deviations. Now ... just when you thought it was safe to do things
> blindly, do you think even statisticians agree on how to compute
> quantiles, like the 90th percentile, for example? It turns out everyone
> agrees on how to compute the 50th percentile -- the median -- but there
> are "opinions" about the others. R has *nine* different ways -- yes,
> that's right -- *nine*. :) If you're either curious or furious, download
> the 13.5 MB R reference manual from
> http://cran.fhcrc.org/doc/manuals/fullrefman.pdf and look up the
> "quantile" function -- it's about 1220 pages into the 2250 page manual. Or
> just install R and type "help(quantile)" at the prompt. :)

   It's not surprising, Ed. These are all concepts without a physical basis.
Therefore they can be defined any way you want. And, each is correct if you
assume certain conditions.

   This is one of the reasons we're having a tough time selling our approach
to quantifying the inherent subjectivity in environmental impact
assessments. No one's done it before, most people have difficulty accepting
that it can be done, and the math can easily glaze the eyes of the listener.
I know because I saw that before I changed my presentation to focus on the
business aspects rather than the math. After all, if it works, that's what
the regulators and companies want; they don't care how, just that it's
technically sound and legally defensible. Which it is. If you're curious,
you can read all about it in my book, "Quantifying Environmental Impact
Assessments Using Fuzzy Logic," published about 18 months ago by
Springer-Verlag.

Rich

-- 
Richard B. Shepard, Ph.D.               |    The Environmental Permitting
Applied Ecosystem Services, Inc.        |          Accelerator(TM)
<http://www.appl-ecosys.com>     Voice: 503-667-4517      Fax: 503-667-8863