Modified Confidence Intervals in Two Variance Components Mixed Model

• 발행기관: 한국자료분석학회
• 저자명: 강관중(Kwan-Joong Kang) 김희재(Hee-Jae Kim) 김규곤(Kyu Kon Kim) 이정형(Jeong-Hyeong Lee)
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.7 No.1, 49~59쪽, 전체 11쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2005.02.28

영문 초록

The two variance components mixed model where is a random vector, is a known matrix with rank , and are unknown parameters. and are mutually independent normal random vectors with zero means, is known matrix. The method of Tukey-Williams confidence intervals for is given by Ofversten(1993). Then, the probability confidence intervals for was given by asymptotic confidence intervals. But we don t know whether they fit exactly the given confidence coefficient 1- or not. From Kang s results, the method of Tukey-Williams probability ranges of confidence intervals for is equal to or greater than 1-. In this case, we can give the method of the better confidence intervals and we can find out which is the better confidence intervals for the given method of Tukey-Williams confidence intervals in the two variance components mixed model.

목차

1. Introduction

2. Two Variance Components Mixed Model

3. The Modified Confidence Intervals

References

키워드

Variance Components Modified Confidence Intervals Two Variance Components Mixed Model Probability Ranges