Un approccio robusto alla verifica d'ipotesi basato sulla funzione di verosomiglianza pesata

Abstract

Weighted version of the Likelihood Ratio, Wald and score tests are proposed for parametric inference. If the parametric model is correct, the Weighted Likelihood tests are asymptotically equivalent to the corresponding Likelihood tests. Breakdown properties of the Weighted Likelihood tests and Likelihood tests are compared. Some comparison with the Hellinger tests (Simpson, 1989) and other robust tests (Heritier e Ronchetti, 1994) are presented too. Examples and Simulations examine the relative performance of the tests in finite samples and real data contest.