On the role played by the fixed bandwidth in the Bickel-Rosenblatt goodness-of-fit test

For the Bickel-Rosenblatt goodness-of-fit test with fixed bandwidth studied by Fan (1998) we derive its Bahadur exact slopes in a neighbourhood of a simple hypothesis f =f0 and we use them to get a better understanding on the role played by the smoothing parameter in the detection of departures from...

Full description

Bibliographic Details
Author: Tenreiro, Carlos
Format: article
Publication Date:2005
Country:España
Institution:Universitat Autònoma de Barcelona
Repository:Dipòsit Digital de Documents de la UAB
Language:English
OAI Identifier:oai:ddd.uab.cat:97450
Online Access:https://ddd.uab.cat/record/97450
Access Level:Open access
Keyword:Goodness-of-fit test
Kernel density estimator
Bahadur efficiency
Contrastos de bondat d'ajust
Estimadors de densitats amb nuclis
Eficiència de Bahadur
Description
Summary:For the Bickel-Rosenblatt goodness-of-fit test with fixed bandwidth studied by Fan (1998) we derive its Bahadur exact slopes in a neighbourhood of a simple hypothesis f =f0 and we use them to get a better understanding on the role played by the smoothing parameter in the detection of departures from the null hypothesis. When f0 is a univariate normal distribution and we take for kernel the standard normal density function, we compute these slopes for a set of Edgeworth alternatives which give us a description of the test properties in terms of the bandwidth h. A simulation study is presented which indicates that finite sample properties are in good accordance with the theoretical properties based on Bahadur local efficiency. Comparisons with the quadratic classical EDF tests lead us to recommend a test based on a combination of bandwidths in alternative to Anderson-Darling or Cramér-von Mises tests.