I have a physical system which can be represented by the following Bayesian network.
It has the following characteristics
The encoded variables are continuous variables
The causal relationships between different variables are equations. For instance,
x5 = f(x1, x2, x3).
In the standard
Bayesian network, we can use
Gibbs sampling or
MCMC to do the forward sampling. For this kind of Bayesian network, how to perform the forward sampling.