Estimating the Parameter of a Exponential Distribution under Multiply Type II Censored Competing Risks Data

• 발행기관: 한국자료분석학회
• 저자명: Kyeongjun Lee Chankeun Park
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.19 No.4, 1711~1720쪽, 전체 10쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2017.08.31

영문 초록

In lifetime data analysis, it is generally note that more than 1 risk factor (RF) may be appear at the same moment. Namely, a failure of unit is often arisen by one of the RFs. Moreover, it is known that the lifetimes of items may not be checked exactly. Therefore, we derive the estimators of parameters of exponential distribution (ED) with multiply type II censored competing risks (MCCR) data. We consider the maximum likelihood estimator (MLE) for parameters of ED with MCCR data. We also obtain the Bayes estimators (BEs) for parameter of the ED with MCCR data under the squared error loss function (SELF), DeGroot loss function (DLF) and precautionary loss function (PLF). Lindley’s approximate (LA) method is used to compute these BEs. The BEs of parameters are better than the respective MLE in terms of mean squared errors (MSEs) and biases. The choice of SELF seems to be a reasonable choice for Bayes estimation of parameters. A real data analysis has been provided.

목차

1. Introduction

2. Maximum Likelihood Estimation

3. Bayes estimation

4. Data Analysis

5. Simulation study

6. Conclusion

키워드

Bayes estimation competing risk exponential distribution Lindley approximation multiply type II censoring