A. Baíllo Moreno, J. Cárcamo Urtiaga
The almost goodness-of-fit test is a procedure to decide if a (parametric) model serves as a good representation of the probability distribution generating the sample. We use the M-estimator of the parameters to determine the approximating model (within the parametric class) to the unknown distribution. The main objective is the approximate validation of a distribution or an entire parametric family up to a pre-specified margin of error. The test statistic is the Lp-distance between the empirical distribution function and the corresponding one of the estimated (parametric) model. The rejection region is determined via an easy-to-implement and flexible bootstrap method.
Palabras clave: Bootstrap; Empirical process; Goodness-of-fit; Model validation
Programado
Estadística no paramétrica I
12 de junio de 2025 15:30
Auditorio 1. Ricard Vinyes