Rainfall data modeling using improved adaptive type-II progressively censored Weibull-exponential samples Authors: Refah Alotaibi, Mazen Nassar, Ahmed Elshahhat Published: 2024 ABSTRACT To get enough data from experiments that last for a long time, a recently unique improved adaptive Type-II progressive censoring technique has been suggested. This study, taking this scheme into consideration, concentrates on some conventional and Bayesian estimation tasks for parameter and reliability indicators, where the underlying distribution is the Weibull-exponential. From a traditional point of view, the likelihood methodology is explored for gaining point and approximate confidence interval estimates. Apart from the standard method, the Bayesian methodology is investigated to obtain the Bayesian point and credible intervals by taking advantage of the Markov chain Monte Carlo technique and the squared error loss function. To differentiate between the traditional and Bayesian estimates, a simulation analysis proceeds under various conditions. In order to put the suggested strategies into application, a pair of rainfall data sets are evaluated and numerous precision criteria are employed to pick the best progressive censoring plan. Introduction In recent times, dozens of lifetime distributions, featuring various forms for the probability density function (PDF) and hazard rate function (HRF), have been offered to boost modelling capability for actual life data sets. Among the ways to achieve this is to add additional parameters to already existing distributions or to combine them. A novel three-parameter Weibull-exponential (WEx) lifetime model was released by Oguntunde et al.1, leveraging the exponential (Ex) distribution as a basis for the model in the Weibull-G family of distributions. The WEx distribution Suppose Y is a lifetime of a product that follows the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {WEx}(\varvec{\psi })$$\end{document}WEx(ψ), where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varvec{\psi }=(\rho ,\xi ,\mu )^\top$$\end{document}ψ=(ρ,ξ,μ)⊤, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho$$\end{document}ρ and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}μ are scale parameters and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\xi$$\end{document}ξ is shape parameter. Let \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\varpi (y;\mu )=e^{\mu y}-1$$\end{document}ϖ(y;μ)=eμy-1, then its PDF and cumulative distribution function (CDF), can be defined as1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} g(y;\varvec{\psi })=\rho \xi \mu [\varpi (y;\mu )]^{\xi -1} e^{\mu y-\rho [\varpi (y;\mu )]^{\xi }},\ y>0,\ \rho ,\xi ,\mu >0 \end{aligned}$$\end{document}g(y;ψ)=ρξμ[ϖ(y;μ)]ξ-1eμy-ρ[ϖ(y;μ)]ξ,y>0,ρ,ξ,μ>0and2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} G(y;\varvec{\psi })=1-e^{-\rho [\varpi (y;\mu )]^{\xi }}, \end{aligned}$$\end{document}G(y;ψ)=1-e-ρ[ϖ(y;μ)]ξ,respectively. The reliability function (RF), and HRF (at a specific time \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$t>0$$\end{document}t>0) of the WEx distribution are, respectively, given by3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} R(t;\varvec{\psi })=e^{-\rho [\varpi (t;\mu )]^{\xi }}\ t>0,\ \rho ,\xi ,\mu >0 \end{aligned}$$\end{document}R(t;ψ)=e-ρ[ϖ(t;μ)]ξt>0,ρ,ξ,μ>0and4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} h(t;\varvec{\psi })=\rho \xi \mu e^{\mu t} [\varpi (t;\mu )]^{\xi -1}. \end{aligned}$$\end{document}h(t;ψ)=ρξμeμt[ϖ(t;μ)]ξ-1.By setting \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho =1$$\end{document}ρ=1, the WEx model returns to the Gompertz distribution with shape parameter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta >0$$\end{document}θ>0. Additionally, by putting \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho =1$$\end{document}ρ=1, the WEx model reduces to the modified Kies exponential distribution by Al-Babtain et al.2. Taking several options of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho$$\end{document}ρ and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\xi$$\end{document}ξ when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu =1$$\end{document}μ=1, several shapes of the density and hazard rate functions of the WEx distribution are depicted in Fig. 1. Figure 1a indicates that the WEx density (1) has a unimodal, bimodal or decreasing shape. Figure 1b shows that the HRF in (4) has a increasing or bathtub shape. Consequently, the WEx model became one of the most useful and applicable distributions for purposes of dependability and data analysis; for instance, based on progressive, hybrid, and adaptive progressive censored samples, Rastogi3, Abushal4, EL-Sagheer et al.5, respectively, studied some estimation issues of the WEx distribution.Fig. 1Shapes of the WEx distribution at \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu =1$$\end{document}μ=1 (a) PDF and (b) HRF. Shapes of the WEx distribution at \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu =1$$\end{document}μ=1 (a) PDF and (b) HRF. Reliability estimation and censoring strategy Since reliability evaluation can yield a quantitative assessment of a product’s durability, it is crucial to life-testing studies. To ensure that the product satisfies stated reliability criteria and to forecast potential shortcomings, this estimation can be very helpful. For a long-lasting reliable product, the investigator can use a portion of data to assess the product’s reliability without wasting time waiting until all of the sample data fails. The literature offers numerous strategies for stopping the experiment before the entire sample fails in order to save both cash and time, see Maswadah6. We call these processes “censorship procedures”. One of the most commonly employed ways for getting data is the Progressive Type-II censoring (TII-PC) scheme. This plan enables the removal of some still-living units at certain prearranged points throughout the trial. For more detail, see Balakrishnan and Cramer7. A different kind of multi-stage censoring was suggested by Ng et al.8 and is referred to as adaptive progressive Type-II censoring (ATII-PC). This strategy resolves the problem of getting a few observed failures in progressive Type-I hybrid censorship plan proposed by Kundu and Joarder9. According to Ng et al.8, the ATII-PC plan performs well in statistical induction when test length is not a significant factor. However, when the test units are incredibly reliable, the testing period will be excessively lengthy, and the APT-IIC will not have worked in guaranteeing an appropriate total test duration. According to Ng et al.8, the APT-IIC performs well in statistical inference when test length is not a significant factor. However, if the test units are incredibly reliable, the testing period will be excessively lengthy, and the APT-IIC will not have succeeded in guaranteeing an appropriate total test duration. Yan et al.10 developed a fresh censoring technique termed an improved adaptive progressive Type-II censoring (ITII-APC) mechanism in order to address this problem. A detailed explanation of the ITII-APC sample is provided as follows: Assume that two boundaries, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{1}0, \end{aligned}$$\end{document}p(ψ)∝ρa1-1ξa2-1μa3-1e-(b1ρ+b2ξ+b3μ),ρ,ξ,μ>0,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{j}, b_{j}>0, j=1,2,3$$\end{document}aj,bj>0,j=1,2,3. The non-informative case can be applied by equating \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{j}$$\end{document}aj and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{j}, j=1,2,3$$\end{document}bj,j=1,2,3 to zero. Adding the observed data acquiring by the LF provided by (6) to the available prior knowledge given by the joint prior distribution in (14), one can derive the joint posterior distribution of the unknown parameters as\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} H(\varvec{\psi }|{{{\underline{\varvec{y}}}}})= & \frac{1}{A} \rho ^{D_{2}+a_{1}-1}\xi ^{D_{2}+a_{2}-1}\mu ^{D_{2}+a_{3}-1}e^{ -(b_{1}\rho +b_{2}\xi +b_{3}\mu )} \nonumber \\ & \times \exp \left\{ \mu \sum _{i=1}^{D_{2}}y_{i}+(\xi -1)\sum _{i=1}^{D_{2}}\log [\varpi (y_{i};\mu )]-\rho Q({{{\underline{\varvec{y}}}}};\mu ,\xi ) \right\} , \end{aligned}$$\end{document}H(ψ|y_)=1AρD2+a1-1ξD2+a2-1μD2+a3-1e-(b1ρ+b2ξ+b3μ)×expμ∑i=1D2yi+(ξ-1)∑i=1D2log[ϖ(yi;μ)]-ρQ(y_;μ,ξ),where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$A=\int _{0}^{\infty }\int _{0}^{\infty }\int _{0}^{\infty }p(\varvec{\psi })L(\varvec{\psi }|{{{\underline{\varvec{y}}}}})d\rho d\xi d\mu$$\end{document}A=∫0∞∫0∞∫0∞p(ψ)L(ψ|y_)dρdξdμ. Utilizing the squared error loss function, the Bayes estimator of the parametric function \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega (\varvec{\psi })$$\end{document}ω(ψ) can be defined as the posterior mean obtained as15\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \tilde{\omega }(\varvec{\psi })=\frac{\int _{0}^{\infty }\int _{0}^{\infty }\int _{0}^{\infty }\omega (\varvec{\psi })p(\varvec{\psi })L(\varvec{\psi }|{{{\underline{\varvec{y}}}}})d\rho d\xi d\mu }{\int _{0}^{\infty }\int _{0}^{\infty }\int _{0}^{\infty }p(\varvec{\psi })L(\varvec{\psi }|{{{\underline{\varvec{y}}}}})d\rho d\xi d\mu }. \end{aligned}$$\end{document}ω~(ψ)=∫0∞∫0∞∫0∞ω(ψ)p(ψ)L(ψ|y_)dρdξdμ∫0∞∫0∞∫0∞p(ψ)L(ψ|y_)dρdξdμ. The squared error loss function is the most commonly used loss function in the literature. Its simplicity makes it particularly suitable for situations where overestimation and underestimation are treated equally. Additionally, it is easy to employ other loss functions, such as LINEX and general entropy, among others. The ratio of integrals in (15) causes the impossibility of getting the Bayes estimator of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega$$\end{document}ω in a closed form. In order to get around this problem, we propose implementing the MCMC technique to obtain both the necessary Bayes estimates and the HPD credible intervals. This subject is covered in the next part. MCMC and Bayes estimates In Bayesian inference,the MCMC techniques are highly effective Monte Carlo methods. It can be used to sample from posterior distributions and calculating posterior values of interest. To make use of the law of large numbers, the MCMC uses repeating random sampling, just as other Monte Carlo approaches. Using this method, the joint posterior distribution must be partitioned into full conditional distributions for each model parameter. A sample has to be obtained from each of these conditional distributions in order to acquire the required estimates. For the unknown parameters \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho , \xi$$\end{document}ρ,ξ, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}μ, the full conditional distributions are given, respectively, by16\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} & H_{1}(\rho |\varvec{\psi }_{-\rho },{{{\underline{\varvec{y}}}}})\propto \rho ^{D_{2}+a_{1}-1}e^{-\rho [Q({{{\underline{\varvec{y}}}}};\mu ,\xi )+b_{1}]}, \end{aligned}$$\end{document}H1(ρ|ψ-ρ,y_)∝ρD2+a1-1e-ρ[Q(y_;μ,ξ)+b1],17\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} & H_{2}(\xi |\varvec{\psi }_{-\xi },{{{\underline{\varvec{y}}}}})\propto \xi ^{D_{2}+a_{2}-1}\exp \left[ \xi \sum _{i=1}^{D_{2}}\log [\varpi (y_{i};\mu )]-\rho Q({{{\underline{\varvec{y}}}}};\mu ,\xi ) -b_{2}\xi \right] \end{aligned}$$\end{document}H2(ξ|ψ-ξ,y_)∝ξD2+a2-1expξ∑i=1D2log[ϖ(yi;μ)]-ρQ(y_;μ,ξ)-b2ξand18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} H_{3}(\mu |\varvec{\psi }_{-\mu },{{{\underline{\varvec{y}}}}})\propto \mu ^{D_{2}+a_{3}-1}\exp \left[ \mu \sum _{i=1}^{D_{2}}y_{i}+(\xi -1)\sum _{i=1}^{D_{2}}\log [\varpi (y_{i};\mu )]-\rho Q({{{\underline{\varvec{y}}}}};\mu ,\xi ) -b_{3}\mu \right] . \end{aligned}$$\end{document}H3(μ|ψ-μ,y_)∝μD2+a3-1expμ∑i=1D2yi+(ξ-1)∑i=1D2log[ϖ(yi;μ)]-ρQ(y_;μ,ξ)-b3μ.One can easily observe that the full conditional distribution of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho$$\end{document}ρ is gamma distribution, with shape parameter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\delta _{1}=(D_{2}+a_{1})$$\end{document}δ1=(D2+a1) and scale parameter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\delta _{1}=[Q({{{\underline{\varvec{y}}}}};\mu ,\xi )+b_{1}]$$\end{document}δ1=[Q(y_;μ,ξ)+b1]. As a result, obtaining samples from (16) with any gamma-generating algorithm is simple. On the other hand, the distributions in (17) and (18) cannot be reduced to any well-known distribution. In this situation, the Metropolis-Hastings (M-H) algorithm is suitable to get the required samples from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{2}(\xi |\varvec{\psi }_{-\xi },{{{\underline{\varvec{y}}}}})$$\end{document}H2(ξ|ψ-ξ,y_) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{3}(\mu |\varvec{\psi }_{-\xi },{{{\underline{\varvec{y}}}}})$$\end{document}H3(μ|ψ-ξ,y_). We can obtain the required samples in accordance with the next steps by using the M-H algorithm with the Gibbs sampling scheme Step 1.Set the initial values as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(\rho ^{(0)},\xi ^{(0)},\mu ^{(0)})=(\hat{\rho },\hat{\xi },\hat{\mu })$$\end{document}(ρ(0),ξ(0),μ(0))=(ρ^,ξ^,μ^).Step 2.Put \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l=1$$\end{document}l=1.Step 3.Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho ^{(l)}$$\end{document}ρ(l) from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Gamma(\delta _{1},\delta _{2})$$\end{document}Gamma(δ1,δ2).Step 4.Simulate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\xi ^{(l)}$$\end{document}ξ(l) using \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{2}(\xi |\varvec{\psi }_{-\xi },{{{\underline{\varvec{y}}}}})$$\end{document}H2(ξ|ψ-ξ,y_) via the M-H algorithm with normal proposal distribution \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N\left( \hat{\xi },{\hat{V}}_{2}\right)$$\end{document}Nξ^,V^2.Step 5.Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu ^{(l)}$$\end{document}μ(l) from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{3}(\mu |\varvec{\psi }_{-\mu },{{{\underline{\varvec{y}}}}})$$\end{document}H3(μ|ψ-μ,y_) using the M-H algorithm with \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N\left( \hat{\mu },{\hat{V}}_{3}\right)$$\end{document}Nμ^,V^3.Step 6.Employ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho ^{(l)},\xi ^{(l)},\mu ^{(l)}$$\end{document}ρ(l),ξ(l),μ(l) to calculate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R^{(l)}$$\end{document}R(l) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h^{(l)}$$\end{document}h(l).Step 7.Replace l by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l+1$$\end{document}l+1.Step 8.Implement the steps 3 to 7, M replications acquire \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \left( \rho ^{(l)},\xi ^{(l)},\mu ^{(l)},R^{(l)},h^{(l)}\right) , l=1,\dots ,M. \end{aligned}$$\end{document}ρ(l),ξ(l),μ(l),R(l),h(l),l=1,⋯,M. Step 1.Set the initial values as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(\rho ^{(0)},\xi ^{(0)},\mu ^{(0)})=(\hat{\rho },\hat{\xi },\hat{\mu })$$\end{document}(ρ(0),ξ(0),μ(0))=(ρ^,ξ^,μ^). Set the initial values as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(\rho ^{(0)},\xi ^{(0)},\mu ^{(0)})=(\hat{\rho },\hat{\xi },\hat{\mu })$$\end{document}(ρ(0),ξ(0),μ(0))=(ρ^,ξ^,μ^). Step 2.Put \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l=1$$\end{document}l=1. Put \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l=1$$\end{document}l=1. Step 3.Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho ^{(l)}$$\end{document}ρ(l) from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Gamma(\delta _{1},\delta _{2})$$\end{document}Gamma(δ1,δ2). Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho ^{(l)}$$\end{document}ρ(l) from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Gamma(\delta _{1},\delta _{2})$$\end{document}Gamma(δ1,δ2). Step 4.Simulate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\xi ^{(l)}$$\end{document}ξ(l) using \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{2}(\xi |\varvec{\psi }_{-\xi },{{{\underline{\varvec{y}}}}})$$\end{document}H2(ξ|ψ-ξ,y_) via the M-H algorithm with normal proposal distribution \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N\left( \hat{\xi },{\hat{V}}_{2}\right)$$\end{document}Nξ^,V^2. Simulate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\xi ^{(l)}$$\end{document}ξ(l) using \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{2}(\xi |\varvec{\psi }_{-\xi },{{{\underline{\varvec{y}}}}})$$\end{document}H2(ξ|ψ-ξ,y_) via the M-H algorithm with normal proposal distribution \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N\left( \hat{\xi },{\hat{V}}_{2}\right)$$\end{document}Nξ^,V^2. Step 5.Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu ^{(l)}$$\end{document}μ(l) from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{3}(\mu |\varvec{\psi }_{-\mu },{{{\underline{\varvec{y}}}}})$$\end{document}H3(μ|ψ-μ,y_) using the M-H algorithm with \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N\left( \hat{\mu },{\hat{V}}_{3}\right)$$\end{document}Nμ^,V^3. Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu ^{(l)}$$\end{document}μ(l) from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{3}(\mu |\varvec{\psi }_{-\mu },{{{\underline{\varvec{y}}}}})$$\end{document}H3(μ|ψ-μ,y_) using the M-H algorithm with \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$N\left( \hat{\mu },{\hat{V}}_{3}\right)$$\end{document}Nμ^,V^3. Step 6.Employ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho ^{(l)},\xi ^{(l)},\mu ^{(l)}$$\end{document}ρ(l),ξ(l),μ(l) to calculate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R^{(l)}$$\end{document}R(l) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h^{(l)}$$\end{document}h(l). Employ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho ^{(l)},\xi ^{(l)},\mu ^{(l)}$$\end{document}ρ(l),ξ(l),μ(l) to calculate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R^{(l)}$$\end{document}R(l) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$h^{(l)}$$\end{document}h(l). Step 7.Replace l by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l+1$$\end{document}l+1. Replace l by \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$l+1$$\end{document}l+1. Step 8.Implement the steps 3 to 7, M replications acquire \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \left( \rho ^{(l)},\xi ^{(l)},\mu ^{(l)},R^{(l)},h^{(l)}\right) , l=1,\dots ,M. \end{aligned}$$\end{document}ρ(l),ξ(l),μ(l),R(l),h(l),l=1,⋯,M. Implement the steps 3 to 7, M replications acquire \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \left( \rho ^{(l)},\xi ^{(l)},\mu ^{(l)},R^{(l)},h^{(l)}\right) , l=1,\dots ,M. \end{aligned}$$\end{document}ρ(l),ξ(l),μ(l),R(l),h(l),l=1,⋯,M. For any parameter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho ,\xi ,\mu ,R(t)$$\end{document}ρ,ξ,μ,R(t) and h(t), say \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega$$\end{document}ω, the Bayes estimate based on the squared error loss function after a suitable burn-in period B can be expressed as given below\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \tilde{\omega }=\frac{1}{M-B}\sum _{l=B+1}^{M}\omega ^{(l)}. \end{aligned}$$\end{document}ω~=1M-B∑l=B+1Mω(l).The HPD credible interval approach provides threshold values from the posterior distribution that indicate an interval containing a specified probability centered around the distribution. This method assumes that all values within the interval are more likely to represent the parameter than any values outside of it. The HPD credible interval is described as the shortest interval that contains a determined percentage of the posterior probability, usually describing the highest-density region. Suppose that \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H^{*}(\omega |{{{\underline{\varvec{y}}}}})$$\end{document}H∗(ω|y_) denotes the posterior distribution of the unknown parameter \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega$$\end{document}ω. Then, the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$100(1-\tau )\%$$\end{document}100(1-τ)% HPD credible interval \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(L^{*},U^{*})$$\end{document}(L∗,U∗) for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega$$\end{document}ω can be obtained by solving the following two equations simultaneously\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \int _{L^{*}}^{U^{*}}H^{*}(\omega |{{{\underline{\varvec{y}}}}})d\omega =1-\tau \end{aligned}$$\end{document}∫L∗U∗H∗(ω|y_)dω=1-τand\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} H^{*}(L^{*}|{{{\underline{\varvec{y}}}}})=H^{*}(U^{*}|{{{\underline{\varvec{y}}}}}). \end{aligned}$$\end{document}H∗(L∗|y_)=H∗(U∗|y_).As it is seen, we cannot obtain the marginal distribution for any of the unknown parameters, which prevents us from theoretically deriving the required HPD credible intervals. Instead, we can adopt the approach proposed by Chen and Shao28 to utilize the generated MCMC samples to obtain these intervals numerically. In order to compute the HPD credible interval, sort \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega ^{(l)}$$\end{document}ω(l) to be \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega _{(l)}, l=B+1,\dots ,M$$\end{document}ω(l),l=B+1,⋯,M. Then, the HPD credible interval of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega$$\end{document}ω is\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \left\{ {\omega _{\left( {l^* } \right) },\omega _{\left( {l^* + \left( {1 - \tau } \right) (M-B) } \right) } }\right\} , \end{aligned}$$\end{document}ωl∗,ωl∗+1-τ(M-B),where\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \omega _{(l^* + [(1-\tau )(M-B)] )} - \omega _{(l^*)} = \mathop {\min }\limits _{1 \leqslant l \leqslant \tau (M-B)} \left\{ {\omega _{( {l + [(1-\tau )(M-B)]} )} - \omega _{(l)} }\right\} , \end{aligned}$$\end{document}ω(l∗+[(1-τ)(M-B)])-ω(l∗)=min1⩽l⩽τ(M-B)ω(l+[(1-τ)(M-B)])-ω(l),at which \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$[\kappa ]$$\end{document}[κ] is determined to be the largest integer that is either less than or equal to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\kappa$$\end{document}κ. Monte Carlo evaluations This section provides various simulations to test the accuracy of the estimates of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho$$\end{document}ρ, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\xi$$\end{document}ξ, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}μ, R(t) , and h(t) proposed in the previous sections. Simulation designs Drawing 1,000 samples from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {WEx}(0.8,0.5,0.2)$$\end{document}WEx(0.8,0.5,0.2), all estimators of the WEx parameters or reliability features are assessed based on different choices of n, m, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\textbf {S}}}$$\end{document}S, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{i},\ i=1,2$$\end{document}Ti,i=1,2. At time \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$t=0.1$$\end{document}t=0.1, the real value of (R(t), h(t) ) is taken as (0.89252,0.57423). Without loss of generality, we take the actual values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho$$\end{document}ρ, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\xi$$\end{document}ξ, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mu$$\end{document}μ as starting points in the all simulation experiments. Besides \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(T_{1},T_{2})=(2,4)$$\end{document}(T1,T2)=(2,4) and (4,6), we consider several values of n, m, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\textbf {S}}}$$\end{document}S such as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n(=40,80)$$\end{document}n(=40,80), m is taken as a failure percent, i.e., \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\frac{m}{n}=50\%$$\end{document}mn=50% and 75%, and three different censoring schemes \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\textbf {S}}}$$\end{document}S namely:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned} \begin{array}{l} \hbox {Scheme-1}: \left( n-m,0^{m-1}\right) ,\\ \hbox {Scheme-2}: \left( 0^{\frac{m}{2}-1},n-m,0^{\frac{m}{2}}\right) ,\\ \hbox {Scheme-3}: \left( 0^{m-1},n-m\right) , \end{array} \end{aligned}$$\end{document}Scheme-1:n-m,0m-1,Scheme-2:0m2-1,n-m,0m2,Scheme-3:0m-1,n-m,where the proposed schemes 1, 2, and 3 represent the left, middle, and right censoring, respectively, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0^{m}$$\end{document}0m means no removals in m stages. To draw an ITII-APC sample from the WEx distribution, do the following steps: Step 1. Set the actual values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {WEx}(\varvec{\psi }).$$\end{document}WEx(ψ) Step 2:Simulate a TII-PC sample as (a)-(d): Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document}α independent observations as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\alpha }_{1}},{{\alpha }_{2}},\dots ,{{\alpha }_{m}}$$\end{document}α1,α2,⋯,αm from uniform distribution.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\gamma }_{i}}=\alpha _{i}^{{{\left( i+\sum \nolimits _{r=m-i+1}^{m}{{{S}_{r}}} \right) }^{-1}}},\ i=1,2,\dots ,m$$\end{document}γi=αi-1,i=1,2,⋯,m.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{u}_{i}}=1-{{\gamma }_{m}}{{\gamma }_{m-1}}\cdots {{\gamma }_{m-i+1}}$$\end{document}ui=1-γmγm-1⋯γm-i+1 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,\dots ,m$$\end{document}i=1,2,⋯,m.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{i}}=\mu ^{-1}\log \left( 1+(-\rho ^{-1}\log (1-{u}_{i}))^{\xi ^{-1}}\right) ,\ i=1,2,\dots ,m$$\end{document}Yi=μ-1log1+(-ρ-1log(1-ui))-1,i=1,2,⋯,m, is the TII-PC sample from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {WEx}(\varvec{\psi })$$\end{document}WEx(ψ). Step 3. Discard the staying sample \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Y_{i},\ i={{J}_{1}}+2,\dots ,m,$$\end{document}Yi,i=J1+2,⋯,m, when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{J}_{1}}$$\end{document}J1 is recorded at \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{T}_{1}}$$\end{document}T1. Step 4. Obtain the first \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m-{{J}_{1}}-1$$\end{document}m-J1-1 order statistics (say \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{{{J}_{1}}+2}},\dots ,{{Y}_{m}}$$\end{document}YJ1+2,⋯,Ym) from a truncated distribution \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\left[ R\left( {{y}_{{{J}_{1}}+1}} \right) \right] }^{-1}{g\left( y\right) }$$\end{document}RyJ1+1-1gy with sample size \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n-{{J}_{1}}-1-\sum \nolimits _{i=1}^{{{J}_{1}}}{{{S}_{i}}}$$\end{document}n-J1-1-∑i=11Si.Step 5. Stop the ITII-APC life-test at: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}$$\end{document}Ym when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}<{{T}_{1}}<{{T}_{2}}$$\end{document}Ym0$$\end{document}t>0), we recommend two statistical packages, namely: The ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {maxLik}$$\end{document}maxLik’ package (by Henningsen and Toomet29) that employs the Newton-Raphson method via ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {maxNR()}$$\end{document}maxNR()’ function in maximizations;The ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {coda}$$\end{document}coda’ package (by Plummer et al.30) that employs the ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$run\_metropolis\_MCMC()$$\end{document}run_metropolis_MCMC()’ function in the Markovian process.Now, we use the M-H algorithm to run a process called MCMC sampler up to 12,000 times and ignore the first 2,000 results. Once the 10,000 MCMC iterations of each WEx parameter are gathered, the Bayes and 95% HPD interval estimates of WEx parameters are obtained. To highlight the performance of the Bayes hyper-parameters, besides the non-informative priors (say, P0), two separate informative sets of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{i},b_{i})$$\end{document}(ai,bi) are used, namely:P0: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{i}=b_{i}=0$$\end{document}ai=bi=0 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3;P1: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{1},a_{2},a_{3})=(4,2.5,1)$$\end{document}(a1,a2,a3)=(4,2.5,1) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{i}=5$$\end{document}bi=5 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3;P2: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{1},a_{2},a_{3})=(8,5,2)$$\end{document}(a1,a2,a3)=(8,5,2) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{i}=10$$\end{document}bi=10 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3.Following the idea presented by Kundu31, the given hyper-parameter values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{i},b_{i},\ i=1,2,3,$$\end{document}ai,bi,i=1,2,3, of the unknown WEx parameters are chosen in such a way that the prior mean matches the expected value of the given actual parameter values. It is better to note here that if one takes P0 into account, the joint posterior density will then be in proportion to the likelihood function (6), i.e., \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H(\varvec{\psi }|{{{\underline{\varvec{y}}}}})\propto (\rho \xi \mu )^{-1} L(\varvec{\psi }|{{{\underline{\varvec{y}}}}})$$\end{document}H(ψ|y_)∝(ρξμ)-1L(ψ|y_). Generally, if the proper prior information is available, it is better to use the informative prior(s) than the non-informative prior(s). Step 1. Set the actual values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {WEx}(\varvec{\psi }).$$\end{document}WEx(ψ) Set the actual values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hbox {WEx}(\varvec{\psi }).$$\end{document}WEx(ψ) Step 2:Simulate a TII-PC sample as (a)-(d): Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document}α independent observations as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\alpha }_{1}},{{\alpha }_{2}},\dots ,{{\alpha }_{m}}$$\end{document}α1,α2,⋯,αm from uniform distribution.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\gamma }_{i}}=\alpha _{i}^{{{\left( i+\sum \nolimits _{r=m-i+1}^{m}{{{S}_{r}}} \right) }^{-1}}},\ i=1,2,\dots ,m$$\end{document}γi=αi-1,i=1,2,⋯,m.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{u}_{i}}=1-{{\gamma }_{m}}{{\gamma }_{m-1}}\cdots {{\gamma }_{m-i+1}}$$\end{document}ui=1-γmγm-1⋯γm-i+1 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,\dots ,m$$\end{document}i=1,2,⋯,m.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{i}}=\mu ^{-1}\log \left( 1+(-\rho ^{-1}\log (1-{u}_{i}))^{\xi ^{-1}}\right) ,\ i=1,2,\dots ,m$$\end{document}Yi=μ-1log1+(-ρ-1log(1-ui))-1,i=1,2,⋯,m, is the TII-PC sample from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {WEx}(\varvec{\psi })$$\end{document}WEx(ψ). Simulate a TII-PC sample as (a)-(d): Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document}α independent observations as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\alpha }_{1}},{{\alpha }_{2}},\dots ,{{\alpha }_{m}}$$\end{document}α1,α2,⋯,αm from uniform distribution.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\gamma }_{i}}=\alpha _{i}^{{{\left( i+\sum \nolimits _{r=m-i+1}^{m}{{{S}_{r}}} \right) }^{-1}}},\ i=1,2,\dots ,m$$\end{document}γi=αi-1,i=1,2,⋯,m.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{u}_{i}}=1-{{\gamma }_{m}}{{\gamma }_{m-1}}\cdots {{\gamma }_{m-i+1}}$$\end{document}ui=1-γmγm-1⋯γm-i+1 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,\dots ,m$$\end{document}i=1,2,⋯,m.Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{i}}=\mu ^{-1}\log \left( 1+(-\rho ^{-1}\log (1-{u}_{i}))^{\xi ^{-1}}\right) ,\ i=1,2,\dots ,m$$\end{document}Yi=μ-1log1+(-ρ-1log(1-ui))-1,i=1,2,⋯,m, is the TII-PC sample from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {WEx}(\varvec{\psi })$$\end{document}WEx(ψ). Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document}α independent observations as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\alpha }_{1}},{{\alpha }_{2}},\dots ,{{\alpha }_{m}}$$\end{document}α1,α2,⋯,αm from uniform distribution. Generate \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document}α independent observations as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\alpha }_{1}},{{\alpha }_{2}},\dots ,{{\alpha }_{m}}$$\end{document}α1,α2,⋯,αm from uniform distribution. Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\gamma }_{i}}=\alpha _{i}^{{{\left( i+\sum \nolimits _{r=m-i+1}^{m}{{{S}_{r}}} \right) }^{-1}}},\ i=1,2,\dots ,m$$\end{document}γi=αi-1,i=1,2,⋯,m. Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{\gamma }_{i}}=\alpha _{i}^{{{\left( i+\sum \nolimits _{r=m-i+1}^{m}{{{S}_{r}}} \right) }^{-1}}},\ i=1,2,\dots ,m$$\end{document}γi=αi-1,i=1,2,⋯,m. Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{u}_{i}}=1-{{\gamma }_{m}}{{\gamma }_{m-1}}\cdots {{\gamma }_{m-i+1}}$$\end{document}ui=1-γmγm-1⋯γm-i+1 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,\dots ,m$$\end{document}i=1,2,⋯,m. Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{u}_{i}}=1-{{\gamma }_{m}}{{\gamma }_{m-1}}\cdots {{\gamma }_{m-i+1}}$$\end{document}ui=1-γmγm-1⋯γm-i+1 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,\dots ,m$$\end{document}i=1,2,⋯,m. Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{i}}=\mu ^{-1}\log \left( 1+(-\rho ^{-1}\log (1-{u}_{i}))^{\xi ^{-1}}\right) ,\ i=1,2,\dots ,m$$\end{document}Yi=μ-1log1+(-ρ-1log(1-ui))-1,i=1,2,⋯,m, is the TII-PC sample from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {WEx}(\varvec{\psi })$$\end{document}WEx(ψ). Set \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{i}}=\mu ^{-1}\log \left( 1+(-\rho ^{-1}\log (1-{u}_{i}))^{\xi ^{-1}}\right) ,\ i=1,2,\dots ,m$$\end{document}Yi=μ-1log1+(-ρ-1log(1-ui))-1,i=1,2,⋯,m, is the TII-PC sample from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {WEx}(\varvec{\psi })$$\end{document}WEx(ψ). Step 3. Discard the staying sample \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Y_{i},\ i={{J}_{1}}+2,\dots ,m,$$\end{document}Yi,i=J1+2,⋯,m, when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{J}_{1}}$$\end{document}J1 is recorded at \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{T}_{1}}$$\end{document}T1. Discard the staying sample \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Y_{i},\ i={{J}_{1}}+2,\dots ,m,$$\end{document}Yi,i=J1+2,⋯,m, when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{J}_{1}}$$\end{document}J1 is recorded at \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{T}_{1}}$$\end{document}T1. Step 4. Obtain the first \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m-{{J}_{1}}-1$$\end{document}m-J1-1 order statistics (say \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{{{J}_{1}}+2}},\dots ,{{Y}_{m}}$$\end{document}YJ1+2,⋯,Ym) from a truncated distribution \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\left[ R\left( {{y}_{{{J}_{1}}+1}} \right) \right] }^{-1}{g\left( y\right) }$$\end{document}RyJ1+1-1gy with sample size \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n-{{J}_{1}}-1-\sum \nolimits _{i=1}^{{{J}_{1}}}{{{S}_{i}}}$$\end{document}n-J1-1-∑i=11Si. Obtain the first \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m-{{J}_{1}}-1$$\end{document}m-J1-1 order statistics (say \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{{{J}_{1}}+2}},\dots ,{{Y}_{m}}$$\end{document}YJ1+2,⋯,Ym) from a truncated distribution \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\left[ R\left( {{y}_{{{J}_{1}}+1}} \right) \right] }^{-1}{g\left( y\right) }$$\end{document}RyJ1+1-1gy with sample size \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n-{{J}_{1}}-1-\sum \nolimits _{i=1}^{{{J}_{1}}}{{{S}_{i}}}$$\end{document}n-J1-1-∑i=11Si. Step 5. Stop the ITII-APC life-test at: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}$$\end{document}Ym when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}<{{T}_{1}}<{{T}_{2}}$$\end{document}Ym0$$\end{document}t>0), we recommend two statistical packages, namely: The ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {maxLik}$$\end{document}maxLik’ package (by Henningsen and Toomet29) that employs the Newton-Raphson method via ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {maxNR()}$$\end{document}maxNR()’ function in maximizations;The ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {coda}$$\end{document}coda’ package (by Plummer et al.30) that employs the ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$run\_metropolis\_MCMC()$$\end{document}run_metropolis_MCMC()’ function in the Markovian process.Now, we use the M-H algorithm to run a process called MCMC sampler up to 12,000 times and ignore the first 2,000 results. Once the 10,000 MCMC iterations of each WEx parameter are gathered, the Bayes and 95% HPD interval estimates of WEx parameters are obtained. To highlight the performance of the Bayes hyper-parameters, besides the non-informative priors (say, P0), two separate informative sets of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{i},b_{i})$$\end{document}(ai,bi) are used, namely:P0: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{i}=b_{i}=0$$\end{document}ai=bi=0 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3;P1: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{1},a_{2},a_{3})=(4,2.5,1)$$\end{document}(a1,a2,a3)=(4,2.5,1) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{i}=5$$\end{document}bi=5 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3;P2: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{1},a_{2},a_{3})=(8,5,2)$$\end{document}(a1,a2,a3)=(8,5,2) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{i}=10$$\end{document}bi=10 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3.Following the idea presented by Kundu31, the given hyper-parameter values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{i},b_{i},\ i=1,2,3,$$\end{document}ai,bi,i=1,2,3, of the unknown WEx parameters are chosen in such a way that the prior mean matches the expected value of the given actual parameter values. It is better to note here that if one takes P0 into account, the joint posterior density will then be in proportion to the likelihood function (6), i.e., \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H(\varvec{\psi }|{{{\underline{\varvec{y}}}}})\propto (\rho \xi \mu )^{-1} L(\varvec{\psi }|{{{\underline{\varvec{y}}}}})$$\end{document}H(ψ|y_)∝(ρξμ)-1L(ψ|y_). Generally, if the proper prior information is available, it is better to use the informative prior(s) than the non-informative prior(s). Stop the ITII-APC life-test at: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}$$\end{document}Ym when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}<{{T}_{1}}<{{T}_{2}}$$\end{document}Ym0$$\end{document}t>0), we recommend two statistical packages, namely: The ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {maxLik}$$\end{document}maxLik’ package (by Henningsen and Toomet29) that employs the Newton-Raphson method via ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {maxNR()}$$\end{document}maxNR()’ function in maximizations;The ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textsf {coda}$$\end{document}coda’ package (by Plummer et al.30) that employs the ‘\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$run\_metropolis\_MCMC()$$\end{document}run_metropolis_MCMC()’ function in the Markovian process.Now, we use the M-H algorithm to run a process called MCMC sampler up to 12,000 times and ignore the first 2,000 results. Once the 10,000 MCMC iterations of each WEx parameter are gathered, the Bayes and 95% HPD interval estimates of WEx parameters are obtained. To highlight the performance of the Bayes hyper-parameters, besides the non-informative priors (say, P0), two separate informative sets of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{i},b_{i})$$\end{document}(ai,bi) are used, namely:P0: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{i}=b_{i}=0$$\end{document}ai=bi=0 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3;P1: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{1},a_{2},a_{3})=(4,2.5,1)$$\end{document}(a1,a2,a3)=(4,2.5,1) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{i}=5$$\end{document}bi=5 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3;P2: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(a_{1},a_{2},a_{3})=(8,5,2)$$\end{document}(a1,a2,a3)=(8,5,2) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$b_{i}=10$$\end{document}bi=10 for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i=1,2,3$$\end{document}i=1,2,3.Following the idea presented by Kundu31, the given hyper-parameter values of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$a_{i},b_{i},\ i=1,2,3,$$\end{document}ai,bi,i=1,2,3, of the unknown WEx parameters are chosen in such a way that the prior mean matches the expected value of the given actual parameter values. It is better to note here that if one takes P0 into account, the joint posterior density will then be in proportion to the likelihood function (6), i.e., \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H(\varvec{\psi }|{{{\underline{\varvec{y}}}}})\propto (\rho \xi \mu )^{-1} L(\varvec{\psi }|{{{\underline{\varvec{y}}}}})$$\end{document}H(ψ|y_)∝(ρξμ)-1L(ψ|y_). Generally, if the proper prior information is available, it is better to use the informative prior(s) than the non-informative prior(s). \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}$$\end{document}Ym when \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${{Y}_{m}}<{{T}_{1}}<{{T}_{2}}$$\end{document}Ym