EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Frequentist and Bayesian Predictive Inference for the Log-Logistic Distribution Under Progressive Type-II Censoring

Ziteng Zhang, Wenhao Gui iD

Entropy · 18.04.2026 · 28 (4) · 466

DOI: 10.3390/e28040466 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 18.04.2026. Verfasst von Ziteng Zhang und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Statistics, Beijing Jiaotong University, Beijing 100044, China. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper investigates the prediction of unobserved future failure times for the heavy-tailed Log-Logistic distribution under Progressive Type-II censoring. We first develop point and interval estimates for the unknown parameters using both frequentist maximum likelihood and Bayesian approaches. For predicting future failures, we derive three distinct point predictors: the Best Unbiased Predictor (BUP), the Conditional Median Predictor (CMP), and the Bayesian Predictor (BP). Corresponding prediction intervals are constructed using frequentist pivotal quantities, Bayesian Equal-Tailed Intervals (ETIs), and Highest Posterior Density (HPD) methods. The Bayesian procedures are implemented via Markov chain Monte Carlo (MCMC) sampling. We evaluate the finite-sample performance of the proposed methodologies through a Monte Carlo simulation study and further validate them using two real-world datasets, namely bladder cancer remission times and guinea pig survival times. The numerical results indicate that the proposed BP, particularly under the empirical prior, provides the most accurate and stable overall performance for point prediction, while the frequentist predictors become less reliable in extreme heavy-tailed settings. For interval prediction, the Bayesian HPD method consistently outperforms the alternatives, substantially reducing interval lengths for right-skewed data while maintaining the nominal coverage probability.

Schlagwörter

Stichwörter der Autor:innen

Log-Logistic distributionMCMC samplingbayes estimationbayes predictionprogressive censoring datasimulation

Autor:innen und Institutionen

  • School of Mathematics and Statistics, Beijing Jiaotong University, Beijing 100044, China.

ORCID-Profile: Wenhao Gui

Zitieren

APA

Zhang, Z., & Gui, W. (2026). Frequentist and Bayesian Predictive Inference for the Log-Logistic Distribution Under Progressive Type-II Censoring. Entropy, 28(4), 466. https://doi.org/10.3390/e28040466

Vancouver

Zhang Z, Gui W. Frequentist and Bayesian Predictive Inference for the Log-Logistic Distribution Under Progressive Type-II Censoring. Entropy. 2026;28(4):466. doi:10.3390/e28040466

BibTeX
@article{Zhang2026,
  author = {Zhang, Ziteng and Gui, Wenhao},
  title = {Frequentist and Bayesian Predictive Inference for the Log-Logistic Distribution Under Progressive Type-II Censoring},
  journal = {Entropy},
  year = {2026},
  volume = {28},
  number = {4},
  pages = {466},
  doi = {10.3390/e28040466}
}
Kostenlos starten →