LARGE-SAMPLE TESTS IN NEGATIVE BINOMIAL DISTRIBUTIONS WITH BOTH PARAMETERS UNKNOWN

by Simos G. Meintanis.

Abstract: We employ the empirical probability generating function in constructing a goodness-of-fit test for negative binomial distributions. The proposed tests, which are in the form of weighted integrals, are shown to be consistent and their asymptotic null distribution is investigated. As the decay of the weight function tents to infinity, limit statistics are obtained. A small simulation study is presented .

Key Words: Probability Generating Function; Goodness-of-Fit; Moment estimation

Author:
Simos G. Meintanis, esimosmei@upatras.gr

Editor: Hector F. Coronel-Brizio,hcoronel@xal.megared.net.mx

READING THE ARTICLE: You can read the article in portable document (.pdf) format (179982 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2682 times since July 24, 2006.


Return to the InterStat Home Page.