I know I can only use the log-likelihoods of two models as selection criterion if they are nested. However, I don't understand this completely. Why isn't it possible to apply this reasoning to non-nested models? I think the higher the log-likelihood the better fits the model the data. I am also aware of other criteria as AIC but this doesn't helps me to understand the issue with the log-likelihoods.
Asked
Active
Viewed 3,014 times
1 Answers
6
You can compare the log-likelihoods of two non-nested models. See Cox (1961), "Tests of separate families of hypotheses"), Proc. Fourth Berkeley Symp. on Math. Statist. and Prob., 1 & Cox (1962), "Further results on tests of separate families of hypotheses", JRSS B, 24, 2. But the log-likelihood ratio test statistic doesn't have the asymptotic chi-square distribution with degrees of freedom equal to the difference in the no. estimated parameters that is so convenient when the models compared are nested. See Wilks (1938), "The large-sample distribution of the likelihood ratio for testing composite hypotheses", Ann. Math. Statist., 9, 1.

Scortchi - Reinstate Monica
- 27,560
- 8
- 81
- 248
-
1If it is comparable, what is needed to be sure to conclude one model is superior? – random_guy Feb 13 '15 at 13:16
-
@Scortchi; The comparability you speak can be intended as follow? if the LLF of model A is greater than one of model B, where the number of parameters of A is equal or lower than B, A is better. This is ok? This my question is related: https://stats.stackexchange.com/questions/546328/distributions-fitting-a-comparison – markowitz Oct 04 '21 at 14:57