Universidad Complutense de Madrid
E-Prints Complutense

Size and power considerations for testing loglinear models using phi-divergence test statistics



Downloads per month over past year

Cressie, Noel A. and Pardo Llorente, Leandro and Pardo Llorente, María del Carmen (2003) Size and power considerations for testing loglinear models using phi-divergence test statistics. Statistica Sinica, 13 (2). pp. 555-570. ISSN 1017-0405


Official URL: http://www3.stat.sinica.edu.tw/statistica/oldpdf/A13n218.pdf


In this article, we assume that categorical data axe distributed according to a multinomial distribution whose probabilities follow a loglinear model. The inference problem we consider is that of hypothesis testing in a loglinear-model setting. The null hypothesis is a composite hypothesis nested within the alternative. Test statistics are chosen from the general class of phi-divergence statistics. This article collects together the operating characteristics of the hypothesis test based on both asymptotic (using large-sample theory) and finite-sample (using a designed simulation study) results. Members of the class of power divergence statistics are compared, and it is found that the Cressie-Read statistic offers an attractive alternative to the Pearson-based and the likelihood ratio-based test statistics, in terms of both exact and asymptotic size and power.

Item Type:Article
Uncontrolled Keywords:Chi-squared distribution; contiguous alternatives; multinomial distribution; nested hypotheses; noncentral chi-squared distribution; powerdivergence statistic.
Subjects:Sciences > Mathematics > Mathematical statistics
ID Code:17833
Deposited On:21 Jan 2013 15:17
Last Modified:26 Feb 2015 09:14

Origin of downloads

Repository Staff Only: item control page