On the determination of the number of outliers in Geometric sample

by Mathachan Pathiyil and E.S Jeevanand.

Abstract: Any one who has carried out experiments and collected data on a number of occasions must have been confronted with outliers, discordant or spurious observations. It is tempting to remove extreme values from a data set because they will incorrectly alter the calculated statistics from the reality. The problem is quite an old one and the main aim is to introduce some degree of objectivity into the rejection of outlying observations. It seems that the problem of identification of the outliers in the geometric set up is not much discussed in the available literature and the present work is an attempt to fill this gap. In this paper we suggest procedures for the determination of the number of outliers present in a sample taken from geometric distribution when the data is in the form of a frequency distribution. We also compare our procedure with identification of outlier procedure based on the posterior density proposed by Kale and Kale and the method of least square by Wu.

Key Words: Geometric distribution, Outliers, Survival function, Bayes method

Authors:
Mathachan Pathiyil, mathachanpathiyil@yahoo.co.in
E.S. Jeevanand, radhajeevanand@yahoo.co.in

Editor: McKean, Joseph W., joe@stat.wmich.edu

READING THE ARTICLE: You can read the article in portable document (.pdf) format (130227 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 2454 times since July 24, 2006.


Return to the InterStat Home Page.