Exact Inference for Competing Risks Model with Generalized Progressive Hybrid Censored Exponential Data

• 발행기관: 한국자료분석학회
• 저자명: Youngseuk Cho Kyeongjun Lee
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.19 No.2, 565~575쪽, 전체 11쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2017.04.30

영문 초록

In this paper, we derive exact likelihood inference for competing risks model (CRM) with generalized progressive hybrid censored exponential data. Based on the maximum likelihood estimators (MLEs) for unknown parameters, we derive the conditional moment generating function (MGF). Also, based on the first moments of the MLEs, we provide the bias adjusted estimators. We compare the proposed estimators by Monte Carlo simulation. For fixed sample and progressive censored sample size, the MSEs decrease as the pre-fixed time increases. For pre-fixed time, sample and progressive censored sample size, the MSEs decrease as the number of guarantee sample size increases. Also, we found that the bias adjusted estimators are better in terms of bias and MSE. But, we found that MLEs are better than the bias adjusted estimators in terms of variances. The simulation step is repeated 10,000 times for the  = 20, 30, and 40 and various generalized progressive hybrid censoring scheme (GPHCS).

목차

1. Introduction

2. Maximum Likelihood Estimation

3. Conditional moment generating function

4. Numerical Experiment

5. Conclusion

키워드

Bias adjusted estimator competing risks model exponential distribution generalized progressive hybrid censoring moment generating function