Timeline for Is it okay to compare fitted distributions with the AIC?
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 10, 2012 at 8:35 | vote | accept | Alexander Engelhardt | ||
| Jul 9, 2012 at 13:40 | comment | added | user10525 | @AlexxHardt This is not quite correct (as I have mentioned in my first comment). Please, have a look at this document about AIC Myths, particularly the last paragraph of pp. 2. Other references by Brian Ripley: 1, 2, 3. | |
| Jul 9, 2012 at 11:21 | comment | added | Alexander Engelhardt | Oh, yes, I agree with the need for penalty. What I don't agree with, is that the likelihood descend from different distributions, i.e. different density families. Aren't we comparing apples with oranges there? I thought the AIC only works for nested models where you add one input variable to your regression and then see how much the likelihood improves. But it would still be the same likelihood family. | |
| Jul 9, 2012 at 10:39 | vote | accept | Alexander Engelhardt | ||
| Jul 9, 2012 at 11:19 | |||||
| Jul 6, 2012 at 13:51 | history | answered | John | CC BY-SA 3.0 |