I'm fairly new to all this so please bear with me...I'm looking at the possibilities attached to valuing optionality (actually in physical assets) using a stochastic mesh. Looking at some of the existing research, the methodology looks feasible (computationally that is over a small time horizon), however I am getting stuck when it comes to determine the transitional density function.My problem is as follows: I have two different, correlated processes upon which the value of the option is calculated. I need to determine, in this example, the density function f(S(t+1)| S(t)=x), where S(t) is a vector-valued process. I understand how the density can be estimated using a kernel where I just have one process, but am having trouble understanding how this gets extended to cope with both of the price processes at the same time.Any help or pointers is very much appreciated.Thanks in advance