Iranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601A Dual Auxiliary Variable Approach to Finite Population Variance Estimation11673595410.22034/jirss.2026.2064841.1120ENAbidHussainDepartment of Statistics, PMAS-Arid Agriculture University, Rawalpindi, Pakistan.0000-0003-4141-0359RabbiaMukhtarDepartment of Statistics, PMAS-Arid Agriculture University, Rawalpindi, Pakistan.0009-0002-2107-7608Muhammad AsimMasoodDepartment of Statistics, PMAS-Arid Agriculture University, Rawalpindi, Pakistan.0000-0002-4856-2411NasirAliDepartment of Statistics, PMAS-Arid Agriculture University, Rawalpindi, Pakistan.0000-0002-0477-6549Journal Article20250705This study proposes a novel estimator for the finite population variance under simple random sampling. The estimator utilizes dual auxiliary information by incorporating the empirical cumulative distribution function (ECDF) of an auxiliary variable. The ECDF, which represents the stochastic process over the unit interval [0,1], is employed to enhance the estimation precision. The performance of the proposed estimator is evaluated through a comprehensive analysis. First, the bias and mean squared error (MSE) of the estimator are derived analytically. Second, a simulation study is conducted to investigate the estimator's behavior under various parametric settings. Finally, an empirical comparison is made with several well-established estimators using five real-world datasets. The results consistently demonstrate the superiority of the proposed estimator in terms of both bias and MSE, suggesting its practical utility.https://jirss.irstat.ir/article_735954_f82178acedb5651bd7905c4e68000488.pdfIranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601Testing Of Symmetry Based On Cumulative Past And Residual Extropy Of Record Values173673717310.22034/jirss.2026.2082667.1175ENSantosh KumarChaudharyDepartment of Statistics, Central University of Jharkhand, Ranchi, India.0000-0002-2350-4266NitinGuptaDepartment of Mathematics, Indian Institute of Technology Kharagpur, India.Journal Article20251231<span class="fontstyle0">This paper proposes new nonparametric tests for symmetry based on cumulative past extropy and cumulative residual extropy of record values, motivated by </span><span class="fontstyle0">a recent characterization of symmetric distributions by </span><span class="fontstyle0">Gupta and Chaudhary </span><span class="fontstyle0">(</span><span class="fontstyle0">2024</span><span class="fontstyle0">). </span><span class="fontstyle0">The proposed estimators are inspired by the methodology introduced by </span><span class="fontstyle0">Vasicek </span><span class="fontstyle0">(</span><span class="fontstyle0">1976</span><span class="fontstyle0">). </span><span class="fontstyle0">The proposed tests do not require estimation of the centre of symmetry, making them </span><span class="fontstyle0">robust and easy to implement. Their asymptotic properties and consistency are established, and critical values are obtained via Monte Carlo simulations. Power is evaluated </span><span class="fontstyle0">under various asymmetric alternatives. Results show that the proposed tests perform </span><span class="fontstyle0">competitively and often outperform existing symmetry tests while maintaining the </span><span class="fontstyle0">nominal significance level. Application of the test to six real-world datasets confirms </span><span class="fontstyle0">its e</span><span class="fontstyle2">ff</span><span class="fontstyle0">ectiveness in detecting symmetric and asymmetric behavior through significant </span><span class="fontstyle0">p-values.</span>https://jirss.irstat.ir/article_737173_87cb53bffd98f175229e1fa7229c5c1b.pdfIranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601Evaluation of the System Lifetime Using a New Mixed δ-Shock Model374773719210.22034/jirss.2026.2048842.1088ENRezaFarhadianDepartment of Statistics, Faculty of Science, Razi University, Kermanshah, Iran.HabibJafariDepartment of Statistics, Faculty of Science, Razi University, Kermanshah, Iran.0000-0001-5191-2796HamedLorvandDepartment of Mathematical Sciences, Isfahan University of Technology, Isfahan, Iran.Journal Article20241224Reliability assessment of complex systems requires accurate modeling of failure mechanisms induced by random external shocks. Classical shock models often focus either on shock magnitudes or on the inter-arrival times between successive shocks, whereas many real-world systems are affected by the joint impact of these two factors. This paper introduces a mixed shock model for systems whose failure is governed by both shock magnitudes and inter-arrival times. In the proposed model, system failure occurs either when the inter-arrival time falls within the interval [α, δ], where 0 ≤ α < δ, or when the magnitude of a single shock exceeds a critical threshold γ. The probability distribution of the system’s stopping time is derived, and the reliability properties of the system’s lifetime are investigated. Numerical examples are provided to illustrate the theoretical findings.https://jirss.irstat.ir/article_737192_744952bb14805f9a7fde00a7f16c09eb.pdfIranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601New Estimation for the Traffic Intensity Parameter of a Single-Server Queueing System with Finite Capacity486473717410.22034/jirss.2026.2080367.1169ENShahram ShahrastaniYaghoobzadehDepartment of Statistics, Payame Noor University, Tehran, Iran0000-0002-8794-2222Journal Article20251209In queueing theory, system is evaluated using performance merics such as the average number of of customers in the queue and system, and the average waiting time. One of the most important parameters for these metrics is the traffic intensity, which must be estimated. This paper proposes a new estimation method and compares it with existing approaches. We focus on the M/M/1/K single-server queueing model ( finite capacity) and estimate the traffic intensity using Bayesian, E-Bayesian, hierarchical Bayesian, and a new EE-Bayesian method. Beacause reducing costs and minimizing customer waiting time are central goales in queueing systems, We consider an estimator suitable if it minimizes average customer waiting time. Using Monte Carlo simulation and a real dataset, We demonstrate the superiority of the proposed method over other estimators.https://jirss.irstat.ir/article_737174_52a412bf72feee5f6c8ce78a03c9475c.pdfIranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601Bayesian Neural Networks for Nonlinear Regression: Posterior Inference, Uncertainty Quantification and Scalability659073719110.22034/jirss.2026.2085011.1182ENFatemehHosseiniSemnan University, Semnan, Iran0000-0003-1826-4090OmidKarimiSemnan University, Semnan, Iran0000-0001-7586-6587Journal Article20260209Bayesian neural networks provide a probabilistic framework for nonlinear regression by combining the expressive flexibility of neural networks with principled uncertainty quantification. However, practical implementation remains challenging because posterior inference is computationally demanding and often requires approximate methods. This study presents a comparative analysis of several inference approaches for Bayesian neural-network regression, including Hamiltonian Monte Carlo, the No- U-Turn Sampler and variational inference, with emphasis on predictive uncertainty, posterior calibration and computational efficiency. The results show that different inference strategies often achieve comparable predictive accuracy, whereas substantially larger differences emerge in uncertainty quantification and scalability. Sampling-based approaches provide more reliable posterior characterization and better-calibrated predictive uncertainty, particularly under complex noise structures, but incur substantially higher computational cost. In contrast, variational inference offers competitive predictive performance together with markedly improved computational efficiency, although it may underestimate posterior uncertainty in more challenging settings. Overall, the findings suggest that the primary practical benefit of Bayesian neural networks lies in reliable and interpretable uncertainty quantification rather than solely improved point prediction. The choice of inference strategy should therefore balance posterior fidelity, uncertainty calibration, and computational scalability according tohttps://jirss.irstat.ir/article_737191_835de59140ff487255adc8a4c99c5a71.pdfIranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601Sampling Scheme of Group Acceptance for Life Testing with the Half Logistic Distribution9110373813610.22034/jirss.2026.2082914.1176ENKarthik NarayanaBoyapatiData Engineer, Master’s in Computer Science, California State University, East Bay, 25800,
Carlos Bee Boulevard, Hayward, CA 94542, USA.Siva Ram PrasadJAssociate Professor, Department of Mathematics, Siddhartha Academy of Higher Education,
Deemed to be University, Vijayawada, Andhra Pradesh, India.Srinivasa RaoBProfessor of Statistics, Department of Mathematics & Humanities, R.V.R. & J.C. College of
Engineering, Chowdavaram, Guntur-522019, Andhra Pradesh, India.0000-0002-6663-436XJournal Article20260105Reliability studies play a crucial role in quality control analysis. Based on these studies, an investigator can reduce both cost and time by efficiently deciding whether to reject or accept the provided lot. Sampling scheme of acceptance deals with the inspection of product lots and the action taken to reject or accept them, and it is one of the earliest schemesapplied in quality control. In present article, a sampling scheme of group acceptance is proposed with truncated life testing under the assumption that item lifetimes follow half logistic distribution. To a specified size of group, the least required count of groups and the corresponding acceptance count are calculated for given buyer’s risk and test completion time. Operating characteristic numbers are computed for different quality levels, and the least ratios of true mean life to stated life corresponding to a given vendor’s risk are calculated. The findings are demonstrated through illustrative examples.https://jirss.irstat.ir/article_738136_c132acabb20ce1ae73a9866b0997a05b.pdfIranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601Handling non-response in the presence of extreme values10412673779210.22034/jirss.2026.1999131.1006ENSyed AbdulRehmanDepartment of Mathematical Sciences, BUITEMS, Pakistan0000-0003-0992-9310ShabbirJavidDepartment of Statistics, Unoversity of Wah, Wah, Pakistan0000-0002-0035-7072Journal Article20230328Non-response and extreme values are two major issues faced by surveyors at the survey and estimation stage; they lead to biased and inefficient estimators. This study proposes four unbiased estimators of the finite population mean in the presence of extreme values when non-response occurs in surveys. We consider two robust ranked set sampling (RSS) designs, called improved paired RSS (IPRSS) and median RSS (MRSS) for the derivation of the proposed estimators. Expressions are provided for the variance of the proposed estimator, and it is mathematically proved that these estimators are more efficient than the Hansen naïve model. Additionally, the efficiency conditions are provided in comparison with the Bouza model of handling non-response in RSS. We conduct a comprehensive simulation study to observe the relative efficiency of the proposed estimators based on hypothetical normal data. For applications in real life, we also consider heart failure data for simulations.https://jirss.irstat.ir/article_737792_18e7a00c66ecc2c50a3dce8b99878d3b.pdfIranian Statistical SocietyJournal of the Iranian Statistical Society1726-405725120260601A novel parametric correlation coefficient for intuitionistic fuzzy numbers12714873815210.22034/jirss.2026.2082115.1173ENZahraBehdaniDepartment of Mathematics and Statistics, Faculty of Energy and Data Sciences, Behbahan Khatam Alanbia University of Technology, Khouzestan, Iran.Journal Article20251226Correlation coefficients are widely used to measure the relationship between variables, but existing approaches for intuitionistic fuzzy numbers (IFNs) often fail to capture negative dependence and lack flexibility. In this paper, we propose a novel parametric correlation coefficient for IFNs based on a newly defined difference operator. The proposed method allows the correlation value to vary within the interval $[-1, 1]$, enabling the representation of both positive and negative relationships. Moreover, by introducing parameters $\alpha$ and $\beta$, the method provides flexibility for decision-makers to evaluate correlations at different levels. Numerical examples and applications in medical diagnosis and multi-criteria decision-making demonstrate the effectiveness, robustness, and practical applicability of the proposed approach. Correlation coefficients are widely used to measure the relationship between variables, but existing approaches for intuitionistic fuzzy numbers (IFNs) often fail to capture negative dependence and lack flexibility.https://jirss.irstat.ir/article_738152_dc3479382dd9378558991bc63f132b8f.pdf