The Clarke test is a likelihood-ratio-based test that can be used for choosing between two non-nested models.
clarke.test(obj1, obj2, sig.lev = 0.05)
It returns a decision.
Objects of the two fitted bivariate non-nested models.
Significance level used for testing.
Maintainer: Giampiero Marra giampiero.marra@ucl.ac.uk
The Clarke (2007) test is a likelihood-ratio-based tests for model selection that use the Kullback-Leibler information criterion, and that can be employed for choosing between two bivariate models which are non-nested.
If the two models are statistically equivalent then the log-likelihood ratios of the
observations should be evenly distributed around zero
and around half of the ratios should be larger than zero. The test follows asymptotically a binomial distribution with
parameters \(n\) and 0.5. Critical values can be obtained as shown in Clarke (2007). Intuitively,
model obj1 is preferred over obj2 if the value of the test
is significantly larger than its expected value under the null hypothesis (\(n/2\)), and vice versa. If
the value is not significantly different from \(n/2\) then obj1 can be thought of as equivalent to obj2.
Clarke K. (2007), A Simple Distribution-Free Test for Non-Nested Model Selection. Political Analysis, 15, 347-363.