Author Topic: Discrete vs Hybrid BN  (Read 115 times)

Offline georgeskerr

  • Newbie
  • *
  • Posts: 1
    • View Profile
Discrete vs Hybrid BN
« on: January 16, 2023, 23:19:23 »
hello
I am trying to assess the performance of two models one is discrete and the other has the same structure but I kept continuous values.
My aim is to assess the sensitivity to findings " Value of information (VOI) " and sensitivity to parameters through the computation of the Posterior Probability Certainty index (through Posterior probability distribution), AUC, and ROC curves.

1.  It is easy to find VOI in the GUI however where and how I can see the posterior probability distribution for the target node?
2. The VOI is found in discrete but not in hybrid networks. So my question what are the common metrics that can be evaluated for both types of networks?
3. Also how are the sensitivity to findings and sensitivity to parameters evaluated in hybrid networks?

I would really appreciate your help. Also if someone has already done this comparison in performance for discrete and hybrid networks would really appreciate their suggestion and help.

Offline Anders L Madsen

  • HUGIN Expert
  • Hero Member
  • *****
  • Posts: 2292
    • View Profile
Re: Discrete vs Hybrid BN
« Reply #1 on: January 19, 2023, 18:40:36 »
Hello,

Here are some quick replies to your questions

  • It is easy to find VOI in the GUI however where and how I can see the posterior probability distribution for the target node?

    Each node has a monitor that displays the posterior probability distribution of the node. You need to compile the network, select the target node, right click on the target node and select toggle monitor window.
  • The VOI is found in discrete but not in hybrid networks. So my question what are the common metrics that can be evaluated for both types of networks?

    Value of information in the GUI does not support hybrid variables.
  • Also how are the sensitivity to findings and sensitivity to parameters evaluated in hybrid networks?

    In hybrid networks there is in many situations not a difference between a discrete observation and a continuous observation. Hence, the documentation applies to both situations. Parameter sensitivity analysis is not supported for the continuous parameters in hybrid networks

I hope this is useful
Anders
HUGIN EXPERT A/S