Expialidocio (X) (asker) Jun 16, 2010:

short-run dispersion Term provided by client. Thanks to all for suggestions!

SJLD Jun 14, 2010:

Have a look at "variation" for "dispersion".

chris collister Jun 14, 2010:

In this particular example (below), the use of the word "instantaneous" is relevant, while "dispersion" here has a rather different meaning, ie the frequency dependence of propagation speed. In your case, "dispersion" refers to the statistical dispersion, ie variability, of whatever is being measured.
http://rose.bris.ac.uk/dspace/bitstream/1983/354/1/fitton_IE...
Note that the meanings of "instant" and "instantaneous" are rather different. For example, my instantaneous speed as measured by a roadside camera might be 40mph (having slowed down for it), while my mean speed over a 2 mile stretch might be 50mph. If my speed is quite constant, then the dispersion in the mph values will be low. On the other hand, if my clutch is dodgy, then the dispersion (ie the standard deviation) in mph values will be high, even though the mean speed may be the same. In terms of how "instantaneous" a measure of instantaneous dispersion (or mean) can be, this discussion could get quite technical...

Expialidocio (X) (asker) Jun 14, 2010:

Sure? Thanks - my discussion entry wasn't a response to yours, just general info. Otherwise, I'm having trouble finding examples of "instantaneous dispersion" or "instant dispersion" or any other variant (distribution, scattering, etc.) that are pertinent. Do you have any references, dictionary or real life? Thanks so much for your contribution so far.

chris collister Jun 14, 2010:

Pretty much the same reasoning applies to dispersion, except that here it's the distribution of the variance (or SD) which alters according to time and circumstance. Translate simply as "instantaneous dispersion".

Expialidocio (X) (asker) Jun 14, 2010:

DISPERSION instantanée Too late to edit question...

chris collister Jun 14, 2010:

The instantaneous distribution depends on when and how the samples are measured. In general the distribution of sample means of a population will tend towards a Normal distribution as more and more measurements are made to determine the population (as opposed to the sample) mean and variance. All this assumes, of course, that the population is in some sense invariant.