How to simulate a Markov chain from the output of two other Markov chains?

Question

I have a Markov chain defined by the output of two other Markov chains as time passes. (i.e., Z = X + Y).

How do I program such a Markov chain in Mathematica?

Answer 1

You can combine processes with TransformedProcess in much the same way that you can combine probability distributions with TransformedDistribution. Here's an example:

proc1 = DiscreteMarkovProcess[{1, 0}, {{0.5, 0.5}, {0.5, 0.5}}];
proc2 = DiscreteMarkovProcess[{1, 0}, {{0.3, 0.7}, {0.3, 0.7}}];
proc3 = TransformedProcess[
   x[t] + y[t],
   {x \[Distributed] proc1, y \[Distributed] proc2},
   t
];
PDF[proc3[3], n] // InputForm
ListPlot @ RandomFunction[
  proc3,
  {0, 10}
]

Piecewise[{{0.15, n == 2}, {0.35, n == 4}, {0.5, n == 3}}, 0]