In this study, we discuss the prediction of k-record values from a future sequence. By considering the Rayleigh distribution, we derive Bayesian point predictions as well as Bayesian prediction intervals for k-record values when the data is collected by the Record ranked set sampling scheme method. At first, based on Squared error and Linex loss functions, the mean squared prediction errors are applied to comparing these point predictors. Next, to evaluate the prediction intervals, the average interval length and coverage probability are computed. Finally, a simulation study and an application example are presented to compare and show the applicability of the paper’s results.
Type of Study:
Original Manuscript |
Subject:
stat Received: 2025/02/7 | Revised: 2026/06/12 | Accepted: 2026/04/14 | Published: 2026/02/26 | ePublished: 2026/02/26