he references for this section are
[Gikhman]
and
[RevuzYor]
.
The process
in
is a Markov process if it is selfdescriptive at any time:


(Markov property)

for some mapping
.
One may always attempt to transform a nonMarkov process into a Markov process
by increasing dimensionality of the process. However, it is not always
possible. For example of such impossibility, think of a diffusion process with
volatility and drift components nontrivially dependent on the entire prior
path.
