Nonparametric Tests Using an Aligned Method for Umbrella Alternatives in a Randomized Block Design

• 발행기관: 한국자료분석학회
• 저자명: Sang Min Seo Dong Hee Kim
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.16 No.1, 1~10쪽, 전체 10쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2014.02.28

영문 초록

In this paper we consider nonparametric tests using an aligned method for umbrella alternatives in a randomized block design with replications. We also assume that there are no interaction effects in the model. The test statistic that we propose is based on Mack, Wolfe (1981)’s statistic, where they have originally suggested testing for umbrella alternatives. Throughout simulation study we compare empirical powers of our proposed statistic with Hettmansperger, Norton (1987)’s statistic and Mack, Skilling (1980)’s statistic. From the simulation results we can show that Hettmansperger, Norton’s statistic has good powers for Cauchy distribution only, but our statistic is wholly efficient for underlying distributions.

목차

1. Introduction

2. The Proposed Test Statistic

3. Simulation Study

4. Conclusions

References

키워드

Umbrella Alternatives Aligned Method Randomized Block Design Nonparametric Test