Distributions of Test Statistics for Outliers in Gamma Samples

• 발행기관: 한국자료분석학회
• 저자명: 김보형(Bo Hyeong Kim) 김동희(Dong Hee Kim)
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.17 No.5, 2291~2300쪽, 전체 10쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2015.10.30

영문 초록

A new statistic U(k) to detect outliers in Gamma distribution is proposed and probability density function of a new statistic U(k using characteristic function is derived. Throughout the simulation study, we have calculated the critical value of the proposed statistic U(k) Dixon (1950) s statistic Dk, Jabbari Nooghabi, Jabbari Nooghabi, Nasiri (2010) s statistic Jk, Kumar, Lalitha (2012) s statistic Zk, Kim, Kim, Park (2012) s statistic Tk. And then, we list the critical values in accordance with the sample sizes, the number of outliers and the scale parameter.

목차

1. Introduction

2. The Proposed Test Statistic

3. Critical Values of the Related Statistics

4. Conclusions and Further Works

References

키워드

The test statistic Outliers Gamma distribution Slippage alternative hypothesis Characteristic function