As I said to David, I was making an analogy which might
be a good way to learn more if followed through. I don't
expect anything to come of it (except my own understanding).
That said, no I didn't have a steady-state model in mind;
I was specifically thinking about trying to formulate a
dynamical model to see if one could obtain any predictive
power for short-term, local price moves. (Again, I don't
buy it; the primary purpose was to give myself a framework
for learning. Any useful results would be a complete
surprise to me.)
At first I had simple analogies in mind
such as price (position), volume (mass), demand (force).
These would have to be set against a background of the
total market and sector movements, as well as continuous
fluctuations. The physical analogy might be a valuable
particles in water subject to ebbing and flowing tides
(market, sector), subject to brownian motion (random local
fluctuations), and on top of that outside forces applied
to particles or classes of simlar particles. Pretty
whimsical, huh? To a first approximation, you would
ignore the "tides" and smooth out the brownian motion
to try to determine the forces at work from the number
of particles moved (volume), from which you might predict
the immediate future movement. Maybe. Or something like
that.
I'm stuck about there at the moment and have not had
a stretch of time at a whack to work at it much beyond
this sort of whimsical stuff. I can already see it's
not right -- doesn't incorporate enough history. Like
I said, the point isn't to be right it's to provide
a framework to learn in. Still, eventually I will impose
it on the thread if anything sensible comes out of it,
even sensible failure <ggg>.
Spots |
