Next: 3: Estimation of
Up: 2: Feefoefum in
Previous: Example of Birth/Death
- One can generalize either the discrete or continuous time cases
by considering a system with again states
and now not
only transitions 
but a general probability 
(discrete) or
(continuous) of transitions from 
to 
at each
trial.
- This theory includes general random walk problem and is discussed
in Volume 1, Feller.
- The mathematics is not very deep---solving recurrence
relations---and the techniques are not useful for the basic problem in
physics (statistics of data reduction).
- So I will not consider further. People with biological bent
should, however, read further in Feller.
- This Markov theory is relevant in discussions of the so called
Metropolis Method in Monte Carlo integration.
Geoffrey Fox, Northeast Parallel Architectures Center at Syracuse University, gcf@npac.syr.edu