Author Topic: Model validation  (Read 13972 times)

Offline ali

  • Newbie
  • *
  • Posts: 3
    • View Profile
Model validation
« on: June 10, 2014, 10:22:42 »
Hi
What do you mean by free parameters in the formula Akaike’s Information Criterion?
Please explain with an example.
To what extent these two criteria have been proposed for the discrete case  of data so that the model is a good model?
Best regards,

Offline Anders L Madsen

  • HUGIN Expert
  • Hero Member
  • *****
  • Posts: 2295
    • View Profile
Re: Model validation
« Reply #1 on: June 14, 2014, 13:59:26 »
The number of free parameters in a discrete CPT is (n-1) * m, where n is the number of states in the child and m is the number of parent configurations.

You can find a lot of information on AIC and BIC by performing a few Google searches.
HUGIN EXPERT A/S

Offline ali

  • Newbie
  • *
  • Posts: 3
    • View Profile
Re: Model validation
« Reply #2 on: June 17, 2014, 06:53:21 »
thank you for your response!
I have trained the  structure model with NPC algorithm, and the variance and mean as follows:
µ          Ϭ
22.5       480.75
34.11   427.27
26.75   97.74
13.87   258.74
118.46   9226
How do you Evaluate the model with AIC=-2870 and BIC=-5325??
Best regards,

Offline Anders L Madsen

  • HUGIN Expert
  • Hero Member
  • *****
  • Posts: 2295
    • View Profile
Re: Model validation
« Reply #3 on: August 02, 2014, 08:18:51 »
You cannot compute the AIC and BIS scores from this information. See section 12.2 of the HUGIN API Reference Manual on how graphical models are scored.

HUGIN EXPERT A/S