Statistical Inference for Heterogeneous Competing Risks Model Under Improved Adaptive Type-II Progressive Censoring.

Area of Science:

• Statistics
• Reliability Engineering
• Survival Analysis

Background:

• Competing risks models are crucial for analyzing multiple failure causes.
• Adaptive Type-II progressive censoring optimizes data collection in reliability studies.
• Statistical inference for complex models requires efficient estimation techniques.

Purpose of the Study:

• To develop statistical inference methods for a heterogeneous competing risks model.
• To implement an improved adaptive Type-II progressive censoring scheme.
• To provide both frequentist and Bayesian estimation approaches for model parameters.

Main Methods:

• Utilized Chen and Weibull distributions for latent lifetimes.
• Developed frequentist methods for point and interval estimation.
• Applied Bayesian inference with Markov Chain Monte Carlo (MCMC) techniques.
• Incorporated approximate confidence intervals, bootstrap confidence intervals, and highest posterior density credible intervals.

Main Results:

• Derived point and interval estimates for model parameters using both frequentist and Bayesian frameworks.
• Demonstrated the effectiveness of the proposed censoring scheme in controlling testing time and ensuring data adequacy.
• Validated the methodologies through extensive Monte Carlo simulations.
• Showcased practical utility via a real-world data application.

Conclusions:

• The proposed statistical inference methods are effective for heterogeneous competing risks models under the specified censoring scheme.
• Both frequentist and Bayesian approaches provide reliable parameter estimates.
• The adaptive censoring scheme balances testing efficiency with data sufficiency for robust analysis.