This simulates the behavior of prices of stocks or other commodities, in a model
sometimes called "geometric brownian motion". The transition from the
price at one time to the next is
Price(t) = Price(t-1) * exp(mu + 0.5 * sigma * Z)
where mu = "drift" , sigma = "volatility"
and Z is the value of a (0,1)-normal random variable.
The choice of seed (and other data) completely determines
the rest, so everything that happens is reproducible.
Note: There are certainly many possible
variations on the formula above, and apparently varying possible
conventions or usages concerning the symbols. I am told by people who
know better than I that there is reason to write, instead,
Price(t) = Price(t-1) * exp((mu-sigma^2/2) + sigma * Z)
But this is not what the applet does, for better or for worse.
© 1997-2004
, Paul Garrett ... [ garrett@math.umn.edu
]
The University of
Minnesota explicitly requires that I state that "The views and
opinions expressed in this page are strictly those of the page
author. The contents of this page have not been reviewed or approved
by the University of Minnesota."