Volume no :17, Issue no: 2, June (2017)

Journal of Statistics: Advances in Theory and Applications

GENERALIZED DIVERGENCE CRITERIA FOR MODEL SELECTION BETWEEN RANDOM WALK AND AR(1) MODEL

Author's: Dhaker Hamza, Papa Ngom, Pierre Mendy and El Hadji Deme
Pages: [83] - [109]
Received Date: May 12, 2017
Submitted by:
DOI: http://dx.doi.org/10.18642/jsata_7100121830

Abstract

We introduce a kernel-type estimators of for continuous distributions. We discuss this approach of goodness-of-fit test for a model selection criterion relative to these divergence measures. Our interest is in the problem to testing for choosing between two models using some informational type statistics (on random walk and autoregressive AR (1)). The limit laws of the estimates and test statistics are given under both the null and the alternative hypotheses. We also describe how to apply estimators and illustrate their efficiency through numerical experiences.

Keywords

divergence measure, kernel estimator, hypothesis testing.