Scholarlink

<p>In this article, we are dealing with two important issues that arise in the competing risks analysis of series system lifetime data. First, we deal with incomplete lifetimes of the system’s components, observed under a Type-I progressive hybrid censoring scheme. Second, we examine situations where the exact cause of failure of any system is unknown. To address these issues, we develop models incorporating cause dependent and time dependent masking probabilities. The Maxwell distribution is considered as the lifetime model for components, and parameter estimation is performed using maximum likelihood and Bayesian approaches. In a simulation study, the derived methodology is explored for varying sample sizes and different censoring patterns under cause and time dependent masking mechanisms. For real-life illustration, the data set of 10,000 hard drives is analyzed. To choose a better model in the presence of various masking options, the predictive power and deviance information criterion are also explored.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Latent Variable Approach to the Analysis of Progressively Hybrid Censored Masked Data

  • 期刊论文

  • 发表日期：

    2025年9月16日

查看全文
  • Sanjeev K Tomer,
  • M S Panwar,
  • Himanshu Rai

摘要

In this article, we are dealing with two important issues that arise in the competing risks analysis of series system lifetime data. First, we deal with incomplete lifetimes of the system’s components, observed under a Type-I progressive hybrid censoring scheme. Second, we examine situations where the exact cause of failure of any system is unknown. To address these issues, we develop models incorporating cause dependent and time dependent masking probabilities. The Maxwell distribution is considered as the lifetime model for components, and parameter estimation is performed using maximum likelihood and Bayesian approaches. In a simulation study, the derived methodology is explored for varying sample sizes and different censoring patterns under cause and time dependent masking mechanisms. For real-life illustration, the data set of 10,000 hard drives is analyzed. To choose a better model in the presence of various masking options, the predictive power and deviance information criterion are also explored.