statistical jargon

[Victor Steinbok]

While I agree with most of what's been written so far, my initial point
is that there are at least two other uses of "spectral analysis" as a
purely mathematical term that has nothing to do with chemistry,
astronomy or physics (although, obviously, it can be used for those
purposes).

Wiki specifically connects the disambiguation of "Spectral analysis" to
Spectral Theory (Banach spaces, etc., in functional analysis) in
mathematics and to Spectral Density Estimation in statistics. Both are
quite distinct from the "pure physical sciences" meaning of spectrum,
which they define internally.

Here's the statistics entry:

> In statistical signal processing, the goal of spectral density
> estimation is to estimate the spectral density (also known as the
> power spectrum) of a random signal from a sequence of time samples of
> the signal. Intuitively speaking, the spectral density characterizes
> the frequency content of the signal. The purpose of estimating the
> spectral density is to detect any periodicities in the data, by
> observing peaks at the frequencies corresponding to these periodicities.

These are not covered under the OED definition at all. So it's not just
a question of emphasis. In Tukey's case, he was one of the inventors of
spectral analysis in statistics, which is why his name shows up in the
quotation in the OED.

There also appears to be a dispute between the Tukey and the [Phil]
Niquette camps as to who coined the term "software". I suppose, that
would extend for its doppelganger "hardware", although, obviously, the
word itself has existed much longer.

     VS-)

On 6/5/2012 8:22 AM, Amy West wrote:
> On 6/5/12 12:00 AM, Automatic digest processor wrote:
>> I think that the problem with this entry is that it gives too much
>> weight to chemical spectral analysis as representing "spectral analysis"
>> as a whole.
>>
>> Pretty much any quantity which can be represented in a frequency domain
>> can have a spectrum, and that spectrum may be analyzed.  Optical or
>> infrared spectra are usually what is being analyzed in chemical spectral
>> analysis (above); much astronomy is done this way.  Audio spectra may be
>> analyzed as well.
> I have to agree with Bill, but based primarily on my own experience: I
> grew up with my dad's astronomical spectrographs lying around the house.
> I didn't learn about chemical spectral analysis (which, ironically, is
> what underlies much of astronomical spectroscopy) until I took chemistry
> in HS and college.
>
> Again, just another point of agreement with Bill: granted that we have
> to give weight to what the earliest cites are talking about, starting
> off with more a more general definition like "analysis of a spectrum: as
> a) analysis of optical or infrared spectra; b) analysis of audio or
> radio spectra. . . ." might be the way to go. I think we have to
> recognize that spectroscopy started off with the visible spectrum, and I
> *think* the cites will show that. . ..
>
> ---Amy West

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org