Two extensions of a parametric model are proposed, each one involving the score function of an alternative parametric model. We show that the encompassing hypothesis is equivalent to standard conditions on the score of each of the extended models. The condition on the first extension gives rise to the standard score encompassing test, while the condition on the second extension induces a so-called reversed score encompassing test. A similar logic is applied to the likelihood ratio, generating a likelihood ratio and a reversed likelihood ratio encompassing test. The ensued test statistics can be based on simulations if certain calculations are too difficult to carry out analytically. We study the first order asymptotic properties of the proposed test statistics under general conditions.
Score test; likelihood ratio test; encompassing; simulation-based inference
In order to set up a list of libraries that you have access to,
you must first login
or sign up.
Then set up a personal list of libraries from your profile page by
clicking on your user name at the top right of any screen.