How to represent moderating variables in a probability model

[sachini]

Hi,

My decision network has random variables, which lead to a certain decision-making task. In addition, there are some moderating variables that affect the association between random variables and the decision nodes.
Can some one let me know how I can represent this kind of a network in GeNiE?

Thanks!!
S

[marek [BayesFusion]]

I will be glad to help but I need to understand what you want precisely. What do you mean that a random variable leads to a certain decision making task? Do the variables influence the value or the decision node? What exactly are the moderating variables? A random variable cannot impact a decision -- decisions are by definition independent of anything. If you know a value of a random variable before making a decision, you can draw an arc between that variable and the decision. A decision node can impact any variable in the model. Please explain what you want to do and I will be glad to help. GeNIe's implementation of influence diagrams is complete and I'm pretty sure that you can model any decision problem (involving discrete variables).
Cheers,

Marek