Timeline for Model selection: can I compare the AIC from models of count data between linear and poisson models?
Current License: CC BY-SA 3.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 29, 2019 at 18:27 | comment | added | robin.datadrivers | So are you saying that you can use criterion like the AIC to compare across two models using different likelihood functions? I've never thought you can do that. Please explain. | |
| Jan 9, 2019 at 20:11 | comment | added | Scortchi♦ | @RichardHardy: I'm not well versed in measure theory but I find 3 obvious: you can't compare a mass with a density. I do agree with your criticism - it's not clear from "different likelihood functions" what kind of difference stops you comparing them. | |
| Jan 9, 2019 at 17:53 | comment | added | Richard Hardy | @Scortchi, The discussion of measures is a little too heavy for me. But here are my takeaways, let us see if they make sense. (1) Calculate the likelihood exactly (do not drop constants) - obvious. (2) Use exactly the same data - obvious to me. (3) Do not mix discrete vs. continuous models - tough. I think (3) is related to (2) in a sense that the likelihood for a continuous model is the likelihood not only for the data points at hand but also for all the unobserved ones in between. Does that make sense? And what is your take on my criticism of the answer here, do you support it? | |
| Jan 9, 2019 at 15:43 | comment | added | Scortchi♦ | @RichardHardy: See Likelihood comparable across different distributions. | |
| Oct 19, 2015 at 18:09 | comment | added | Richard Hardy | I am not sure I agree with your first sentence. I must admit I do not know all the assumptions behind the AIC but surely the models do not have to be in the same class of models to be comparable. The first thing to look at is that the response variable is exactly the same (not a transformation, not another sample), but other than that it is not so trivial. Here is a related discussion, unfortunately without a very detailed answer. | |
| Aug 21, 2015 at 7:57 | comment | added | Vincent Laufer | @robin.datadrivers - does the same logic mean that comparing AIC between poisson and negative binomial models would be meaningless? the likelihood functions are similar but slightly different... | |
| Mar 4, 2015 at 13:00 | vote | accept | tomka | ||
| Feb 25, 2015 at 14:56 | history | edited | robin.datadrivers | CC BY-SA 3.0 |
added 194 characters in body
|
| Feb 25, 2015 at 14:41 | history | answered | robin.datadrivers | CC BY-SA 3.0 |