Research Output
Articles
Publication Date: 2025
Biometrical Journal (03233847)67(2)
The mean residual life (MRL) function plays an important role in the summary and analysis of survival data. The main advantage of this function is that it summarizes the information in units of time instead of a probability scale, which requires careful interpretation. Ranked set sampling (RSS) is a sampling technique designed for situations, where obtaining precise measurements of sample units is expensive or difficult, but ranking them without referring to their accurate values is cost-effective or easy. However, the practical application of RSS is hindered because each sample unit is required to assign a unique rank. To alleviate this difficulty, Frey developed a novel variation of RSS, called RSS-t, that records and utilizes the tie structure in the ranking process. In this paper, we propose several different nonparametric estimators for the MRL function based on RSS-t. Then, we compare the proposed estimators with their counterparts in simple random sampling (SRS) and RSS, where tie information is not utilized. We also implemented our proposed estimators on a real data set related to patient waiting times for liver transplantation, to show their applicability and efficiency in practice. Our results show that using ties information leads to an improved statistical inference for the MRL function, and therefore a smaller sample size is needed to reach a predetermined precision. © 2025 Wiley-VCH GmbH.
Publication Date: 2024
International Journal of Biostatistics (15574679)20(2)pp. 571-583
The mean residual lifetime (MRL) of a unit in a population at a given time t, is the average remaining lifetime among those population units still alive at the time t. In some applications, it is reasonable to assume that MRL function is a decreasing function over time. Thus, one natural way to improve the estimation of MRL function is to use this assumption in estimation process. In this paper, we develop an MRL estimator in ranked set sampling (RSS) which, enjoys the monotonicity property. We prove that it is a strongly uniformly consistent estimator of true MRL function. We also show that the asymptotic distribution of the introduced estimator is the same as the empirical one, and therefore the novel estimator is obtained "free of charge", at least in an asymptotic sense. We then compare the proposed estimator with its competitors in RSS and simple random sampling (SRS) using Monte Carlo simulation. Our simulation results confirm the superiority of the proposed procedure for finite sample sizes. Finally, a real dataset from the Surveillance, Epidemiology and End Results (SEER) program of the US National Cancer Institute (NCI) is used to show that the introduced technique can provide more accurate estimates for the average remaining lifetime of patients with breast cancer. © 2024 Walter de Gruyter GmbH, Berlin/Boston.
Publication Date: 2024
Scandinavian Journal of Statistics (14679469)51(2)pp. 447-484
This study intends to introduce kernel mean embedding of probability measures over infinite-dimensional separable Hilbert spaces induced by functional response statistical models. The embedded function represents the concentration of probability measures in small open neighborhoods, which identifies a pseudo-likelihood and fosters a rich framework for statistical inference. Utilizing Maximum Mean Discrepancy, we devise new tests in functional response models. The performance of new derived tests is evaluated against competitors in three major problems in functional data analysis including Function-on-Scalar regression, functional one-way ANOVA, and equality of covariance operators. © 2023 The Board of the Foundation of the Scandinavian Journal of Statistics.
Publication Date: 2023
Statistical Papers (09325026)64(1)pp. 161-177
The mean past lifetime (MPL) is an important tool in reliability and survival analysis for measuring the average time elapsed since the occurrence of an event, under the condition that the event has occurred before a specific time t> 0. This article develops a nonparametric estimator for MPL based on observations collected according to ranked set sampling (RSS) design. It is shown that the proposed estimator is a strongly uniform consistent estimator of MPL. It is also proved that the introduced estimator tends to a Gaussian process under some mild conditions. A Monte Carlo simulation study is employed to evaluate the performance of the proposed estimator with its competitor in simple random sampling (SRS). Our findings show the introduced estimator is more efficient than its counterpart estimator in SRS as long as the quality of ranking is better than random. Finally, an illustrative example is provided to describe the potential application of the developed estimator in assessing the average time between the infection and diagnosis in HIV patients. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.
Publication Date: 2023
Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability (17480078)237(6)pp. 1100-1113
In the literature of system reliability and other fields related to the time to occurrence of an event, shock models play an important role. In this paper, we assume that a system is subject to shocks that occur according to a counting process describing the number of shocks that arrive during a specified time interval. As the magnitude of the damage imposed to the system by each shock is a crucial parameter to the system’s survival, we investigate some important random variables related to this parameter. A random variable of interest associated with this process is the first time, after a pre-specified time (Formula presented.), at which the amount of a shock damage to the system gets greater than the maximum of damages imposed to the system until time (Formula presented.). We obtain the reliability function of this random variable and investigate various properties of it and some other related random variables. In order to explore further the results, we examine two commonly used processes in the literature, that is, the non-homogeneous Poisson process and the Pólya process. © IMechE 2022.
