One of the simplest stochastic processes is an ordinary
random walk, a process
whose increments
from time to time
take exclusively the values +1 or -1. Additionally, we assume the
increments to be i.i.d. and independent of the starting value
. Hence, the ordinary random walk can be written as:
|
(5.1) |
independent and
for all
Letting the process go up by and go down by , instead, we
obtain a more general class of binomial
processes:
für alle
where and are constant (=up,
=down).
Linear interpolation of the points reflects the time
evolution of the process and is called a path of an ordinary
random walk. Starting in , the process moves on the grid of
points
Up to time , can grow at most
up to (if
) or can fall at least to
(if
). Three paths of an ordinary
random walk are shown in Figure 4.1 (),
4.2 () and Figure 4.3
().
Fig.:
Three paths of a symmetric ordinary random walk. ()-intervals
around the drift (which is zero) are given as well.
SFEBinomp.xpl
|
Fig.:
Three paths of an ordinary random walk with . ()-intervals around
the drift (which is the line with negative slope) are given as well.
SFEBinomp.xpl
|
Fig.:
Three paths of an ordinary random walk with . ()-intervals around
the drift (which is the line with positive slope) are given as well.
SFEBinomp.xpl
|
For generalized binomial processes the grid of possible paths is
more complicated. The values which the process starting in
can possibly take up to time are given by
where
If, from time 0 to time , the process goes up times and
goes down times then
That
is, of increments
take the value ,
and increments take the value . The grid of possible paths
is also called a binomial tree.
The mean of the symmetric ordinary random walk
starting
in 0 () is for all times equal to 0 :
Otherwise, the random walk has a trend or
drift, for
it has a
positive drift and for
it has a negative
drift. The process grows or falls in average:
since it holds for all increments
Hence,
the trend is linear in time. It is the upward sloping line in
Figure 4.3 () and the downward sloping line
in Figure 4.2 ().
For the generalized binomial process with general starting
value it holds analogously
and
thus:
As time evolves the set of values grows, and its variability
increases. Since the summands in (4.1) are independent
and
for all , the variance of
is given by (refer to Section 3.4):
Hence, the variance of grows linearly with time. So does the
standard deviation. For the random walks depicted in Figure
4.1 (), Figure 4.2 ()
and Figure 4.3 () the intervals
are shown as well. For
large , these intervals should contain 95% of the
realizations of processes.
The variance of the increments can be easily computed. We use the
following result which holds for the binomial distribution. Define
or
|
(5.2) |
we obtain the following representation of the binomial process
|
(5.3) |
where
|
(5.4) |
is a distributed random variable.
Given the distribution of , the distribution of is
specified for all . It can be derived by means of a simple
transformation of the binomial distribution . From
equations (4.2) to (4.4) we obtain for
:
and for large the distribution of can be approximated
by:
For
, the following
approximation holds for
:
Figure 4.4 shows the fit for .
Fig.:
The distribution of 100 paths of an ordinary symmetric random walk
of length and a kernel density estimation of 100 normally
distributed random variables.
SFEbinomv.xpl
|