IJPAM: Volume 35, No. 1 (2007)

GOODNESS-OF-FIT TESTS VIA PARTITIONS OF
CHI-SQUARED TEST STATISTICS DERIVED
FROM CHARACTERIZING CONDITIONS

Dominik Szynal
Institute of Mathematics
Maria Curie-Sk\lodowska University
Pl. M. Curie-Sk\lodowskiej 1
Lublin, PL-20-031, POLAND
e-mail: szynal@golem.umcs.lublin.pl


Abstract.There are presented goodness-of-fit tests via conditions characterizing continuous distributions in terms of moments of record values. We study partitions of the chi-squared test statistics for testing goodness-of-fit into two components, each asymptotically distributed as a chi-squared variate with one degree of freedom. The formulae of the tests obtained in this way are remarkably simple and they are expressed in original sample values or in terms of extremal statistics.

Received: February 7, 2007

AMS Subject Classification: 62E10, 62G10, 62G30, 62G32

Key Words and Phrases: order statistics, record values, moments, exponential, inverse exponential, normal, inverse Gaussian distributions, characterizations, goodness-of-fit tests, powers

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2007
Volume: 35
Issue: 1