Research article Special Issues

Optimizing analgesic pain relief time analysis through Bayesian and non-Bayesian approaches to new right truncated Fréchet-inverted Weibull distribution

  • Received: 11 September 2023 Revised: 05 November 2023 Accepted: 09 November 2023 Published: 23 November 2023
  • MSC : 62N05, 62N02, 62N01, 62H12, 62F15, 62F10, 62F40

  • This research introduces a novel right-truncated distribution, termed the right truncated Fréchet-inverted Weibull distribution, and elucidates its mathematical properties including density, cumulative, survival and hazard functions. Various statistical attributes such as moments, quantile, mode and moment-generating functions are explored. These properties indicate the efficiency in modeling pain relief time for patients and the number of recoveries of Leukemia patients. Furthermore, estimation techniques, including maximum likelihood and Bayesian methods, are applied to progressive type-Ⅱ right-censored samples to derive parameter estimation of the proposed distribution. Asymptotic properties are employed to approximate confidence intervals for both reliability and hazard functions. Bayesian estimates are refined using both symmetric and asymmetric loss functions. The suitability of the proposed estimation methodologies is validated through simulation studies. The theoretical framework is applied to two real-world lifetime data sets, thereby substantiating their practical utility in medical areas.

    Citation: Nora Nader, Dina A. Ramadan, Hanan Haj Ahmad, M. A. El-Damcese, B. S. El-Desouky. Optimizing analgesic pain relief time analysis through Bayesian and non-Bayesian approaches to new right truncated Fréchet-inverted Weibull distribution[J]. AIMS Mathematics, 2023, 8(12): 31217-31245. doi: 10.3934/math.20231598

    Related Papers:

    [1] Samah M. Ahmed, Abdelfattah Mustafa . Estimation of the coefficients of variation for inverse power Lomax distribution. AIMS Mathematics, 2024, 9(12): 33423-33441. doi: 10.3934/math.20241595
    [2] Refah Alotaibi, Mazen Nassar, Zareen A. Khan, Ahmed Elshahhat . Analysis of Weibull progressively first-failure censored data with beta-binomial removals. AIMS Mathematics, 2024, 9(9): 24109-24142. doi: 10.3934/math.20241172
    [3] Hassan Okasha, Mazen Nassar, Saeed A. Dobbah . E-Bayesian estimation of Burr Type XII model based on adaptive Type-Ⅱ progressive hybrid censored data. AIMS Mathematics, 2021, 6(4): 4173-4196. doi: 10.3934/math.2021247
    [4] Xue Hu, Haiping Ren . Statistical inference of the stress-strength reliability for inverse Weibull distribution under an adaptive progressive type-Ⅱ censored sample. AIMS Mathematics, 2023, 8(12): 28465-28487. doi: 10.3934/math.20231457
    [5] Refah Alotaibi, Mazen Nassar, Zareen A. Khan, Ahmed Elshahhat . Statistical analysis of stress–strength in a newly inverted Chen model from adaptive progressive type-Ⅱ censoring and modelling on light-emitting diodes and pump motors. AIMS Mathematics, 2024, 9(12): 34311-34355. doi: 10.3934/math.20241635
    [6] Ahmed Elshahhat, Refah Alotaibi, Mazen Nassar . Statistical inference of the Birnbaum-Saunders model using adaptive progressively hybrid censored data and its applications. AIMS Mathematics, 2024, 9(5): 11092-11121. doi: 10.3934/math.2024544
    [7] Haiping Ren, Xue Hu . Estimation for inverse Weibull distribution under progressive type-Ⅱ censoring scheme. AIMS Mathematics, 2023, 8(10): 22808-22829. doi: 10.3934/math.20231162
    [8] M. Nagy, H. M. Barakat, M. A. Alawady, I. A. Husseiny, A. F. Alrasheedi, T. S. Taher, A. H. Mansi, M. O. Mohamed . Inference and other aspects for $ q- $Weibull distribution via generalized order statistics with applications to medical datasets. AIMS Mathematics, 2024, 9(4): 8311-8338. doi: 10.3934/math.2024404
    [9] Ahmed R. El-Saeed, Ahmed T. Ramadan, Najwan Alsadat, Hanan Alohali, Ahlam H. Tolba . Analysis of progressive Type-Ⅱ censoring schemes for generalized power unit half-logistic geometric distribution. AIMS Mathematics, 2023, 8(12): 30846-30874. doi: 10.3934/math.20231577
    [10] A. M. Abd El-Raheem, Ehab M. Almetwally, M. S. Mohamed, E. H. Hafez . Accelerated life tests for modified Kies exponential lifetime distribution: binomial removal, transformers turn insulation application and numerical results. AIMS Mathematics, 2021, 6(5): 5222-5255. doi: 10.3934/math.2021310
  • This research introduces a novel right-truncated distribution, termed the right truncated Fréchet-inverted Weibull distribution, and elucidates its mathematical properties including density, cumulative, survival and hazard functions. Various statistical attributes such as moments, quantile, mode and moment-generating functions are explored. These properties indicate the efficiency in modeling pain relief time for patients and the number of recoveries of Leukemia patients. Furthermore, estimation techniques, including maximum likelihood and Bayesian methods, are applied to progressive type-Ⅱ right-censored samples to derive parameter estimation of the proposed distribution. Asymptotic properties are employed to approximate confidence intervals for both reliability and hazard functions. Bayesian estimates are refined using both symmetric and asymmetric loss functions. The suitability of the proposed estimation methodologies is validated through simulation studies. The theoretical framework is applied to two real-world lifetime data sets, thereby substantiating their practical utility in medical areas.



    The family of probability distributions based on the Weibull distribution has been extensively used in numerous research papers. The Weibull distribution (WD) is a continuous probability distribution that was first studied by Fréchet [1] to describe the distribution of particle volume. Recently, it was noticed that the inverse Weibull distribution (IWD) is a very important distribution used in reliability analysis, such as, the depreciation of mechanical components, e.g., pistons and engines, see Balakrishnan[2].

    Assume X is a random variable that follows the inverse Weibull distribution with two parameters, then the cumulative distribution function (CDF) and the probability density function (pdf) for x0 defined as

    Q(x;α,β)=eαxβ,x0,α>0,β>0 (1.1)

    and

    q(x;α,β)=αβx(β+1)eαxβ,x0,α>0,β>0, (1.2)

    respectively, where α and β are the scale and shape parameters.

    Corderio and Lemonte [3] proposed the parameter estimates for the inverse Weibull distribution using the maximum likelihood and least squares methods.

    When we decrease the original distribution's domain to a smaller one, we get truncated distributions, which are conditional distributions. When it is impossible to determine what is happening above or below the phenomenon being examined, truncation is used. So, by setting the appropriate truncated value to the upper whisker limit, we employ a technique to exceed the outliers in the data. Behdani et al. [4] acknowledged that the Pareto distribution is a well-known long-tailed distribution and proposed a truncated variant to circumvent its long tail. Zaninetti [5] proposed a right and left truncated gamma distribution that serves as the distribution's upper and lower boundaries. The distribution's properties and features were studied. El-Din et al. [6] derived the probability density function of the mid-truncated Lindley distribution as a particular case. The failure of the components and units that make up an operating system's main structure was researched by statisticians in the fields of industrial and mechanical engineering. They employed statistical inference techniques to collect the data for their study, which also included examining the lifetime of the operational units until failure and predicting the reliability and risk functions for the entire system using data. The number of experimental units and the length of the lifetime experiment of some units must be limited because they are expensive and extremely valuable. Due to the time and cost constraints of the experiment, it is frequently necessary to remove items before they fail. In such cases, it is not possible for an experimenter to collect all of the sample data in a short amount of time during a lifetime test. This brings us to the subject of censoring, where some operating units are eliminated from the experiment due to a particular censoring scheme, see Aggarwala and Balakrishnan [7]. The progressive type-Ⅱ technique supports the demands of observing good estimators with no lifetime experimental lack. As the examiner assigns independent and identical units on the test, the progressive type-Ⅱ scheme can be observed and summarized as follows: Assume that n independent items are placed through a test with X1,X2,,Xn failure times that are continuous and identically distributed. Assume that a censoring scheme (R1,R2,,Rm) is predetermined so that immediately after the first failure X1,R1 surviving items are removed from the experiment at random, and immediately after the second failure X2,R2 surviving items are removed from the experiment at random. When the mth observed failure occurs, the remaining Rm surviving items are removed from the test, or Xm. The order statistics of size m that have been gradually type Ⅱ right-censored from a sample of size n using a progressive censoring scheme (R1,R2,,Rm) are the m ordered observed failure times represented by X(R1,R2,,Rm)1:m:n,X(R1,R2,,Rm)2:m:n,,X(R1,R2,,Rm)m:m:n. There is no doubt that n=m+R1+R2++Rm. The particular situation of traditional type Ⅱ right-censored sampling occurs when R1=R2==Rm1=0 and R=nm. The progressive censoring scheme has been studied by many researchers, such as [8].

    Many studies, like Meeker et al. [9], Mann et al. [10], Lawless [11], and Buzaridah et al. [12] have explored progressive type-Ⅱ censoring with varied failure time breakdowns. Balakrishnan [2] investigated the characteristics of progressively censored samples and reported that a collection of several advances in the inferential procedures depend on types-Ⅰ and -Ⅱ censored samples. A generalization of the progressive type-Ⅱ censoring was obtained, and it has more realistic applications, for example, Montanari et al. [13], as well as by Bairamov and Eryilmaz [14]. Additionally, Mahmoud et al. [15] presented a lifetime parameter estimation for a new extension of the exponential distribution with a mechanical application. Ramadan et al. [16] introduced the Bayesian estimation of some distributions under progressive type-Ⅱ censoring. Haj Ahmad et al. [17,18] used different models under a progressive type-Ⅱ censoring scheme and modeled water reservoir capacity, and the strength of aircraft's windows respectively. Shrahili et al. [19] derived some recurrence relations for the moments. Luo et al. [20] introduce modeling and estimation of a reliability system with a dynamic functioning area and some lifetime constraints. More work was achieved by many authors, for example, Buzaridah et al. [21] among others. This study's objective is to compare the two methods of inference with a new right-truncated Fréchet-inverted Weibull (RTFIW) distribution based on the progressive type-Ⅱ censoring scheme with an application.

    The basic objective of creating these models is to truncate the original distribution's domain to a smaller one. Additionally, based on complete data, the asymptotic confidence intervals for the maximum likelihood estimator (MLE) of the unknown parameters were examined. In addition, the Markov chain Monte Carlo (MCMC) technique was derived for the Bayesian estimator. The purpose of this study is to present both traditional and Bayesian estimating methods for the unknown parameters of the RTFIW distribution under the progressive type-Ⅱ censoring scheme, where Bayesian methods are based on two different types of loss functions. In addition, a simulation study is conducted to evaluate the accuracy of the results and the quality of the loss function choice.

    The contents of this paper are organized as follows: The structural characteristics of the RTFIW distribution are introduced in Section 2. In Section 3, we introduce order statistics, extreme values, the quantile function, the moments, and the moment-generating function. In Section 4, asymptotic confidence intervals based on the MLE are computed along with the point estimation of the parameters, the hazard rate, and reliability functions. In Section 5, methods are detailed to estimate the unknown parameters using Bayes estimation under the square error and the linear exponential (LINEX) loss function estimators. A simulation study is undertaken in Section 6, to evaluate the effectiveness of the different estimation methods developed in this study. In Section 7, two real data are examined. Finally, concluding remarks are summarized in Section 8.

    The right truncated Fréchet-inverted Weibull (RTFIW) distribution has a CDF F(x) that can be obtained by

    F(x;α,β,b)=Q(x;α,β)Q(b;α,β)=eα(xβbβ),0<xb,α>0,β>0. (2.1)

    If the cumulative distribution function of a random variable X is given by Eq (2.1), it can be given as

    F(x;α,β,b)=eα(xβbβ),0<xb,α>0,β>0, (2.2)

    where α and β are the scale and shape parameters respectively. The value of b is a constant that represents the truncated part.

    The probability density function pdf for RTFIW is given by

    f(x;α,β,b)=ddxF(x)=αβxβ+1eα(xβbβ),0<xb,α>0,β>0. (2.3)

    The survival function S(x) and the hazard rate function h(t) for the RTFIW distribution are given by

    S(x;α,β,b)=P(X>x)=1F(x;α,β,b)=1eα(xβbβ),0<xb,α>0,β>0 (2.4)

    and,

    h(x;α,β,b)=f(x;α,β,b)S(x;α,β,b)=αβeα(xβbβ)xβ+1(1eα(xβbβ)),0<xb,α>0,β>0, (2.5)

    respectively.

    Figures 14 depict the RTFIW distribution's density function, the cumulative distribution function, hazard rate, and survival function for various parameter choices. Figure 1 shows the behavior of the pdf when changing the parameters α,β where b=4. Figure 2 shows how the CDF behaves, affected by the change in parameters α,β where b=4. The behavior of CDF is increasing depending on the values of the parameters. Figure 3 demonstrates the behavior of S(x) when the parameters α,β are changed, where b=4, and how it decreases based on the parameter values. Figure 4 depicts how changing the settings affects the behavior of h(x). The behavior of h(x) may exhibit unimodality. Depending on the values of the parameters, it is either increasing or decreasing.

    Figure 1.  The pdf of the RTFIW with different parameter values.
    Figure 2.  The CDF of the RTFIW with different parameter values.
    Figure 3.  The reliability function of the RTFIW for different parameter values.
    Figure 4.  The hazard function of the RTFIW for different parameter values.

    Now, we will propose several statistical characteristics of the RTFIW distribution.

    The Quantile function has a variety of significant applications, including calculating the median, skewness, and kurtosis, as well as generating random variables, see Moors [22]. The quantile function of X is given by

    x=Q(z)=F1(z)=F1(eα(xβbβ)), (3.1)

    where zuniform(0,1), it is simple to obtain random numbers using

    x=(1bβ1αlog(z))1β. (3.2)

    By inputting z=0.5, the median of the RTEIW distribution from Eq (3.2) is defined as

    median=(1bβ1αlog(0.5))1β. (3.3)

    By differentiating the probability density function of the RTEIW distribution at x>0 and making it equal to zero. The mode is defined as:

    ddx(αβxβ+1eα(xβbβ))=0

    and

    αβeαbβ[(β1)xβ2eαxβ+αβx2β2eαxβ]=0. (3.4)

    The moments of distributions are used to characterize several of the most essential model properties, such as dispersion, skewness, and kurtosis. As a result, the rth moment of the RTFIW distribution is given by

    μr=E(xr)=b0xrf(x;α,β,b)dx, (3.5)
    μr=b0xrαβxβ+1eα(xβbβ)dx
    =αβeαbβb0xrβ1eαxβdx.

    Let αxβ=t, we get

    μr=αβeαbβαbβ0(1αt)rβ1et(1α)1β1βt1β1dt =αβeαbβαbβ0(1α)rβ11β(t)rβ11β1etdt, (3.6)

    where γ(rβ1β1,αbβ) is the lower incomplete gamma function. Then,

    μr=αβeαbβ(αbβ)rβ1β1Γ(rβ1β1)eαbβk=0(αbβ)kΓ(rβ1β1+αbβ)=αβ(αbβ)rβ1β1Γ(rβ1β1)k=0(αbβ)kΓ(rβ1β+αbβ). (3.7)

    The moment generating function (MGF) of the RTFIW distribution is the following:

    MX(t)=0etxf(x;α,β,b)dx. (3.8)

    As a result, using Taylor series to extend MX(t) produces

    MX(t)=b0αβxβ+1etxeα(xβbβ)dx=αβr=0b0trxrr!xβ1eα(xβbβ)dx=r=0trr!μr. (3.9)

    Hence the MGF of the RTFIWD is

    MX(t)=r=0trr!αβ(αbβ)rβ1β1Γ(rβ1β1)k=0(αbβ)kΓ(rβ1β+αbβ).

    The fields of reliability and testing provide several applications for order statistics. Numerous areas of statistical reasoning include the use of order statistics. Let X1, X2,....Xn represent a random sample of size n from the RTFIW with cumulative distribution function F(x) and probability density function f(x). Following that, order statistics are represented by the notations X (1)X(2)....X(n), X(1)=min(X1,X2,....Xn), and X(n)=max(X1,X2,....Xn). The pdf of the RTFIW's kth-order statistics is

    f(k:n)(x)=n!(k1)!(nk)![αβxβ+1eα(xβbβ)]×[eα(xβbβ)]k1×[1eα(xβbβ)]nk.

    The pdf of f(1:n)(x)-order statistics when k=1 is

    f(1:n)(x)=n[αβxβ+1eα(xβbβ)]×[1eα(xβbβ)]n1.

    The pdf of f(n:n)(x)-order statistics when k=n is

    f(n:n)(x)=n[αβxβ+1eα(xβbβ)]×[eα(xβbβ)]n1.

    Distribution of maximum, minimum, and median

    Assume that X1,X2,...,Xn are samples selected at random from a continuous distribution with PDF and CDF. Now we will compute the PDF for X(1), the smallest value of the sample. In order statistics, it is usually easier to begin by considering the CDF. Then, the pdf for the minimum value from distribution is given by

    fX(1)(x)=n[1eα(xβbβ)]n1[αβxβ+1eα(xβbβ)]. (3.10)

    Then, the pdf for the maximum and median from the RTFIW distribution is given by

    fX(n)(x)=n[eα(xβbβ)]n1[αβxβ+1eα(xβbβ)] (3.11)

    and

    fmed(x)=(2m+1)!m!m!(eα(xβbβ))m(1eα(xβbβ))m[αβxβ+1eα(xβbβ)]. (3.12)

    According to a progressive type-Ⅱ censored sample, see [23], with the RTFIW failures times X1:m:n,X2:m:n,....,Xm:m:n,1mn and the removal units are R1,R2,...,Rm, m<n, the likelihood function is expressed by

    L(α,β,bx)=Cmi=1f(xi:m:n;α,β)[1F(xi:m:n;α,β)]Ri,m<n, (4.1)

    where C=n(nR11)....(nm1i=1(Ri1)), is a constant.

    The likelihood function follows as

    L(α,β,bx)=Cmi=1(αβxβ+1i:m:neα(xβi:m:nbβ))(1eα(xβi:m:nbβ))Ri=Cαmβmmi=1eα(xβi:m:nbβ)mi=1xβ+1i:m:nmi=1(1eα(xβi:m:nbβ))Ri. (4.2)

    The log-likelihood function is obtained as

    mlogα+mlogβmi=1α(xβi:m:nbβ)(β+1)mi=1log(xi:m:n)+mi=1Rilog(1eα(xβi:m:nbβ)). (4.3)

    Notice that b is a constant. By computing the partial derivatives of with respect to α and β and equating each one to zero, we get,

    α=mαmi=1(xβi:m:nbβ)+mi=1Ri(xi:m:neα(xβi:m:nbβ)bβeα(xβi:m:nbβ)1eα(xβi:m:nbβ)) (4.4)

    and

    β=mβmi=1log(xi:m:n)mi=1α(bβlog(b)log(xi:m:n)xβi:m:n)+mi=1Ri[αbβeα(xβi:m:nbβ)log(b)αlog(xi:m:n)xi:m:neα(xβi:m:nbβ)1eα(xβi:m:nbβ)]. (4.5)

    The estimates for Eqs (4.4) and (4.5) without a closed-form solution are obtained using the Newton-Raphson iteration method. Additionally, by substituting α and β with ˆα and ˆβ as shown, the MLEs of S(t) and h(t) can be determined using the invariance property of MLEs. The entries of the Fisher information matrix Iij= E{[2(Ψ)/ψiψj]}, where i,j=1,2 and Ψ=(ψ1,ψ2)=(α,β), represent the asymptotic variances and covariances of the MLEs, ˆα and ˆβ. Consequently, it is challenging to find the precise closed forms for the aforementioned demands. The observed Fisher information matrix ˆIij= {[2(Ψ)/ψiψj]}Ψ=ˆΨ, which is produced by eliminating the expectation operator E, which will be used to generate confidence intervals (CIs) for the parameters. This is represented by

    ˆI(ˆα,ˆβ)=(2α22αβ2βα2β2)(α,β)=(ˆα,ˆβ), (4.6)

    where

    2α2=mα2+mi=1[Ri(xβi:m:n+bβ)2eα(xβi:m:nbβ)1eα(xβi:m:nbβ)Ri(xβi:m:n+bβ)2(eα(xβi:m:nbβ))2(1eα(xβi:m:nbβ))2] (4.7)
    2β2=mβ2+mαbβlog(b)2mi=1αxβi:m:nlog(xi:m:n)2+mi=1[Riα(xβi:m:nlog(xi:m:n)2bβlog(b)2)eα(xβi:m:nbβ)1eα(xβi:m:nbβ)Riα2(xβi:m:nlog(xi:m:n)+bβlog(b))2eα(xβi:m:nbβ)1eα(xβi:m:nbβ)Riα2(xβi:m:nlog(xi:m:n)+bβlog(b))2(eα(xβi:m:nbβ))2(1eα(xβi:m:nbβ))2] (4.8)
    2αβ=mbβlog(b)+mi=1xβi:m:nlog(xi:m:n)+mi=1[Ri(xβi:m:nlog(xi:m:n)+bβlog(b))eα(xβi:m:nbβ)1eα(xβi:m:nbβ)+Riα(xβi:m:nlog(xi:m:n)+bβlog(b))(xβi:m:n+bβ)eα(xβi:m:nbβ)1eα(xβi:m:nbβ)+Riα(xβi:m:nlog(xi:m:n)+bβlog(b))(xβi:m:n+bβ)(eα(xβi:m:nbβ))2(1eα(xβi:m:nbβ))2] (4.9)

    and

    2βα=mbβlog(b)+mi=1xβi:m:nlog(xi:m:n)+mi=1[Ri(xβi:m:nlog(xi:m:n)+bβlog(b))eα(xβi:m:nbβ)1eα(xβi:m:nbβ)+Riα(xβi:m:nlog(xi:m:n)+bβlog(b))(xβi:m:n+bβ)eα(xβi:m:nbβ)1eα(xβi:m:nbβ)+Riα(xβi:m:nlog(xi:m:n)+bβlog(b))(xβi:m:n+bβ)(eα(xβi:m:nbβ))2(1eα(xβi:m:nbβ))2]. (4.10a)

    The observed information matrix ˆI(α,β) is inverted in order to derive the approximate asymptotic variance-covariance matrix [ˆV] for the MLEs, therefore, it is defined as

    [ˆV]=ˆI1(ˆα,ˆβ)=(Var(ˆα)cov(ˆα,ˆβ)cov(ˆβ,ˆα)Var(ˆβ)). (4.11)

    A multivariate normal distribution with a mean (α,β) and covariance matrix I1(α,β) is believed to be similar to the Normal distribution under certain regularity conditions, see Lawless [11]. The two-sided 100(1γ)% confidence interval of α and β and can be obtained by

    ˆα±Zγ2Var(ˆα) (4.12)

    and

    ˆβ±Zγ2Var(ˆβ),

    where Zγ2 represents the right-tail probability γ2 of the standard normal distribution.

    It is also important to determine their variances in order to calculate the asymptotic CI of S(t) and h(t), which are functions in the parameters α and β. The approximation of estimation of the variance of ˆS(t) and ˆh(t) are computed using the delta approach, see Greene [24]. The variance of ˆS(t) and ˆh(t) can be estimated using this method, respectively, by

    ˆσ2ˆS(t)=[ˆS(t)]T[ˆV][ˆS(t)]and ˆσ2ˆh(t)=[ˆh(t)]T[ˆV][ˆh(t)],

    where the gradient of ˆS(t)and ˆh(t) with respect to αand β are denoted by ˆS(t)and ˆh(t), respectively. Then, the (1γ)100% two-sided confidence intervals of S(t) and h(t) can be provided by

    ˆS(t)±Zγ2ˆσ2ˆS(t) and ˆh(t)±Zγ2ˆσ2ˆh(t). (4.13)

    The Bayesian estimation of the RTFIW distribution based on progressive type-Ⅱ censoring is obtained. According to Zellner et al., [25], the following gamma prior distributions are applied when α and β are independent parameters.

    {π1(α)=αa11exp(b1α),                 α>0,π2(β)=βa21exp(b2β),                  β>0,          (5.1)

    where, the hyper-parameters ai and bi, i=1,2 are non negative and known. Combining the likelihood function from Eq (4.3) and the priors from Eq (5.1) to get the posterior distribution of the parameters α and β, which is represented by π(α,βx)

    π(α,βx)π1(α)π2(β)L(α,βx)00π1(α)π2(β)L(α,βx)dαdβ. (5.2)

    The square error loss function (SEL) is a symmetry-identical loss function that is frequently used. It is defined as the squared difference between the actual parameter ϕ and an estimator ˆϕ.

    L(ϕ,ˆϕ)=(ˆϕϕ)2. (5.3)

    As a result, under the SEL function the Bayes estimator of any function of α and β, say g(α,β), can be calculated as follows

    ˆgBS(α,βx)=Eα,βx_(g(α,β)). (5.4)

    Hence,

    Eα,βx_(g(α,β))=00g(α,β)π1(α)π2(β)L(α,βx_)dαdβ00π1(α)π2(β)L(α,βx_)dαdβ. (5.5)

    For a parameter ϕ, the LINEX loss function (L()) is denoted by

    L()=(ecc1), c0, =ˆϕϕ. (5.6)

    LINEX loss is used to discuss Bayesian estimation and prediction. As a result, under the LINEX loss function, the Bayes estimate of a function g(α,β) is

    ˆgBL(α,βx_)=1clog[E(ecg(α,β)x_)],c0, (5.7)

    where

    E(ecg(α,β)x_)=00ecg(α,β)π1(α)π2(β)L(α,βx_)dαdβ00π1(α)π2(β)L(α,βx_)dαdβ. (5.8)

    where the calculations in Eqs (5.5) and (5.8) cannot be evaluated analytically. Thus, in Eq (5.2), the Markov chain Monte Carlo MCMC technique is employed to produce samples from the joint posterior density function. The Metropolis-Hasting (M-H) and Gibbs sampling methods are two popular MCMC approaches in statistics. From Eq (5.2), the joint posterior distribution is

    π(α,βx)αm+a11βm+a21eb1αb2βmi=1xβ+1i:m:nmi=1eα(xβi:m:nbβ)mi=1(1eα(xβi:m:nbβ))Ri. (5.9)

    The posterior conditionals for α and β are

    π1(αβ,x)αm+a11eb1αmi=1eα(xβi:m:nbβ)mi=1(1eα(xβi:m:nbβ))Ri (5.10)

    and

    π2(βα,x)βm+a21eb2βmi=1x(β+1)i:m:nmi=1eα(xβi:m:nbβ)mi=1(1eα(xβi:m:nbβ))Ri. (5.11)

    However, the conditional posteriors of α and β in Eqs (5.10) and (5.11) do not have standard forms. Consequently, it is difficult to sample directly by standard methods; however, the plots presented in Figures 5 and 6 display that they have curves that are symmetric and hence, similar to the normal distribution.

    Figure 5.  The conditional posterior density of MCMC results of the α parameter.
    Figure 6.  The conditional posterior density of MCMC results of the β parameter.

    As a result, Gibbs sampling is not an obvious choice, meaning the Metropolis-Hastings sampler is necessary for the MCMC approach implementations. Within Gibbs sampling, the Metropolis-Hastings method is as follows:

    (1) Begin by including an accurate assumption (α(0),β(0)).

    (2) Let k=1.

    (3) Create α(k) and β(k) by using the M-H technique from π1(α(k1)β(k),x_) and π2(β(k1)α(k),x_) with standard proposed distributions N(α(k1),var(α)) and N(β(k1),var(β)), where var(α) and var(β) are the variance of parameters α and β respectively.

    (4) Create a proposed α from N(α(k1),var(α)) and β from N(β(k1),var(α)).

    (i) Compute the possibilities of acceptance.

    ωα=min[1,π1(αβ(k1),x_)π1(α(j1)β(k1),x_)],ωβ=min[1,π2(βα(k),x_)π2(β(k1)α(k),x_)],}

    (ii) Produce a Z1 and Z2 from a uniform distribution.

    (iii) If Z1<ωα, accept the suggestion and put α(k)=α, else put α(k)=α(k1).

    (iv) If Z2<ωβ, accept the suggestion and put β(k)=β, else put β(k)=β(k1).

    (5) Evaluate h(k)(t) and S(k)(t).

    (6) k=k+1.

    (7) We can get α(l),β(l),S(l)(t) and h(l)(t),l=1,2,...N by repeating steps 3 to 6, N times.

    (8) The 100(1γ)% CRIs of ψj are (ψj(Nγ/2),ψj(N(1γ/2))), that computes the CRs of α, β, S(t) and h(t) ψ(l)j,j=1,2,3,4,(ψ1,ψ2,ψ3,ψ4)=(α,β,S(t),h(t)) as ψ(1)j<ψ(2)j...<ψ(N)j.

    The chosen samples are ψ(i)j,k=M+1,...N, for sufficiently large N. The Bayesian estimations of ψj based on the SEL function are given by

    ˆψj=1NMNk=M+1ψ(k). (5.12)

    The Bayesian estimations ofψk from Eq (5.8) under the LINEX loss function are

    ˆψj=1clog[1NMNk=M+1ecφ(k)],j=1,2,3,4. (5.13)

    For each simulation, 1000 progressive type-Ⅱ censored samples are used to evaluate parameter estimators, some lifetime parameters, and the RTFIW distribution's hazard and reliability functions. To create a sample size of progressive type-Ⅱ censored from the RTFIW distribution with initial values of α = 0.5 and β = 0.05. The mean square error (MSE) of the estimators of the results α,β,S(t) and h(t), at t=0.4, has been estimated based on a comparison of the various methods. The various approaches are used to estimate the 95% confidence intervals CIs derived from asymptotic distributions of the MLEs and CRIs. The coverage probability (CP) and the average confidence interval lengths (ACLs) are developed to assess the performance of estimation methods under study. The percentage of times the interval contains the interest-generating beginning value is known as a coverage probability (CP). In this study, the following progressive schemes are introduced:

    ● Scheme Ⅰ: R1=nm,Rj=0for j1.

    ● Scheme Ⅱ: Rm2=Rm2+1=nm2,Rj=0 for jm2 and jm2+1.

    ● Scheme Ⅲ: Rm=nm,Rj=0for jm.

    The results of the estimated parameters and their MSE are presented in Tables 14. Tables 5 and 6 present the results of the ACL and CP of the 95% CIs.

    Table 1.  MSE of ML and Bayesian estimates of α with α = 0.5.
    LINEX
    (n, m) CS MLE SEL c = - 2 c = 0.0001 c = 2
    (30, 20) 0.3853 0.3843 0.3843 0.3842 0.3843
    (0.0727) (0.0739) (0.0739) (0.0738) (0.0737)
    0.4046 0.4036 0.4037 0.4036 0.4036
    (0.063) (0.0626) (0.0626) (0.0625) (0.0624)
    0.4238 0.4237 0.4236 0.4237 0.4236
    (0.641) (0.064) (0.064) (0.063) (0.0639)
    (40, 15) 0.3794 0.3793 0.3794 0.3793 0.3792
    (0.0783) (0.0789) (0.079) (0.0785) (0.0784)
    0.4256 0.4264 0.4265 0.4264 0.4263
    (0.0768) (0.767) (0.0767) (0.0767) (0.766)
    0.4564 0.4567 0.4569 0.4567 0.4565
    (0.071) (0.0719) (0.72) (0.0716) (0.714)
    (40, 20) 0.3784 0.3781 0.3781 0.3780 0.3779
    (0.0689) (0.0686) (0.0689) (0.0685) (0.0684)
    0.3988 0.3989 0.399 0.43988 0.3987
    (0.0636) (0.0639) (0.0639) (0.0631) (0.0630)
    0.4364 0.4363 0.4364 0.4363 0.4362
    (0.0608) (0.0613) (0.0613) (0.0611) (0.0610)
    (50, 15) 0.3817 0.3815 0.3814 0.3810 0.3809
    (0.0759) (0.0748) (0.0759) (0.0749) (0.0741)
    0.4338 0.4344 0.4346 0.4344 0.4342
    (0.081) (0.0813) (0.0814) (0.0810) (0.0807)
    0.4499 0.4496 0.4497 0.4496 0.4495
    (0.0683) (0.0682) (0.0683) (0.0680) (0.0677)
    (60, 15) 0.406 0.4058 0.406 0.4058 0.4057
    (0.0883) (0.0889) (0.0889) (0.0884) (0.0880)
    0.4198 0.4197 0.4197 0.4196 0.4195
    (0.0783) (0.0785) (0.0786) (0.0784) (0.0779)
    0.4474 0.448 0.4482 0.4481 0.4477
    (0.0673) (0.0684) (0.0686) (0.0683) (0.0680)
    (60, 20) 0.407 0.4059 0.407 0.4059 0.4058
    (0.0884) (0.0890) (0.0890) (0.0885) (0.0881)
    0.4199 0.4198 0.4198 0.4197 0.4196
    (0.0784) (0.0784) (0.0787) (0.0785) (0.0780)
    0.4475 0.449 0.4483 0.4482 0.4478
    (0.0675) (0.0686) (0.0687) (0.0684) (0.0683)

     | Show Table
    DownLoad: CSV
    Table 2.  MSE of MLE and Bayesian estimates of β with β = 0.05.
    LINEX
    (n, m) CS MLE SEL c = - 2 c = 0.0001 c = 2
    (30, 20) 0.0633 0.0632 0.0632 0.0631 0.0630
    (0.0008) (0.0006) (0.0008) (0.0005) (0.0004)
    0.0614 0.0613 0.0614 0.0612 0.0611
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0003)
    0.06 0.06 0.063 0.062 0.060
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0002)
    (40, 15) 0.0634 0.0633 0.0634 0.0633 0.0631
    (0.0007) (0.0006) (0.0006) (0.0005) (0.0004)
    0.0608 0.0607 0.0607 0.0606 0.0605
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0003)
    0.0582 0.0581 0.0581 0.0580 0.0579
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0002)
    (40, 20) 0.0632 0.0628 0.0628 0.0627 0.0625
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0003)
    0.0617 0.0614 0.0614 0.0613 0.0612
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0002)
    0.0586 0.0584 0.0584 0.0583 0.0582
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    (50, 15) 0.063 0.062 0.062 0.061 0.060
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0003)
    0.0598 0.0597 0.0597 0.40596 0.0595
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0003)
    0.0577 0.0577 0.0577 0.0576 0.0574
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    (60, 15) 0.0608 0.0607 0.0607 0.0606 0.0675
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    0.0608 0.0607 0.0607 0.0606 0.40605
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0002)
    0.0578 0.40579 0.0579 0.0578 0.0577
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0002)
    (60, 20) 0.0609 0.0608 0.0608 0.0607 0.06076
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0002)
    0.0609 0.40608 0.40608 0.0607 0.0606
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0003)
    0.0579 0.0580 0.0580 0.0579 0.0578
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0003)

     | Show Table
    DownLoad: CSV
    Table 3.  MSE of ML and Bayesian estimates of S(t) with t = 0.4.
    LINEX
    (n, m) CS MLE SEL c = - 2 c = 0.0001 c = 2
    (30, 20) 0.0457 0.0455 0.0455 0.0454 0.0453
    (0.0003) (0.0004) (0.0004) (0.0003) (0.0002)
    0.0475 0.0474 0.0474 0.0472 0.0471
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    0.0489 0.0488 0.0488 0.0487 0.0486
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    (40, 15) 0.0449 0.0448 0.0448 0.0447 0.0446
    (0.005) (0.004) (0.004) (0.003) (0.002)
    0.0479 0.0478 0.0478 0.0477 0.0476
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0001)
    0.0505 0.0504 0.0504 0.0503 0.0502
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0001)
    (40, 20) 0.0455 0.0454 0.0454 0.0453 0.0452
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    0.048 0.047 0.047 0.046 0.045
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    0.0495 0.0494 0.0494 0.0493 0.0492
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    (50, 15) 0.0446 0.0445 0.0445 0.0444 0.0443
    (0.0004) (0.0004) (0.0004) (0.0004) (0.0004)
    0.048 0.0479 0.0479 0.0478 0.0477
    (0.0006) (0.0005) (0.0005) (0.0004) (0.0002)
    0.0499 0.0498 0.0498 0.0497 0.0495
    (0.0006) (0.0004) (0.0004) (0.0002) (0.0001)
    (60, 15) 0.0457 0.0456 0.0456 0.0455 0.0453
    (0.0006) (0.0006) (0.0004) (0.0003) (0.0002)
    0.0473 0.0472 0.0472 0.0471 0.0470
    (0.0007) (0.0006) (0.0006) (0.0004) (0.0003)
    0.0497 0.0496 0.0496 0.0495 0.0494
    (0.0004) (0.0003) (0.0003) (0.0002) (0.0001)
    (60, 20) 0.0458 0.0457 0.0457 0.0456 0.0454
    (0.0007) (0.0006) (0.0006) (0.0004) (0.0002)
    0.0474 0.0473 0.0473 0.0472 0.0471
    (0.0008) (0.0007) (0.0007) (0.0005) (0.0004)
    0.0498 0.0497 0.0497 0.0496 0.0495
    (0.0005) (0.0004) (0.0004) (0.0003) (0.0002)

     | Show Table
    DownLoad: CSV
    Table 4.  MSE of ML and Bayesian estimates of h(t) with t = 0.4.
    LINEX
    (n, m) CS MLE SEL c = - 2 c = 0.0001 c = 2
    (30, 20) 1.1289 1.1288 1.1288 1.1287 1.1286
    (0.0019) (0.0016) (0.0016) (0.0014) (0.0012)
    1.1253 1.1252 1.1252 1.1251 1.1250
    (0.0015) (0.0013) (0.0013) (0.0012) (0.0011)
    1.1229 1.1229 1.1229 1.1227 1.1226
    (0.0013) (0.0012) (0.0012) (0.0011) (0.0010)
    (40, 15) 1.1295 1.1294 1.1294 1.1293 1.1292
    (0.0017) (0.0016) (0.0016) (0.0014) (0.0013)
    1.1242 1.1241 1.1241 1.1240 1.1239
    (0.0015) (0.0013) (0.0013) (0.0011) (0.0010)
    1.1193 1.1192 1.1192 1.1191 1.1190
    (0.0011) (0.0009) (0.0009) (0.0008) (0.0007)
    (40, 20) 1.1283 1.1282 1.1282 1.1281 1.1280
    (0.0015) (0.0014) (0.0014) (0.0013) (0.0011)
    1.1256 1.1255 1.1255 1.1254 1.1252
    (0.0013) (0.0012) (0.0012) (0.0011) (0.0010)
    1.1203 1.1202 1.1202 1.1201 1.1200
    (0.004) (0.003) (0.003) (0.002) (0.001)
    (50, 15) 1.1279 1.1278 1.1278 1.1277 1.1276
    (0.0016) (0.0015) (0.0015) (0.0014) (0.0012)
    1.1229 1.1228 1.1228 1.1227 1.1226
    (0.0019) (0.0017) (0.0017) (0.0016) (0.0014)
    1.1192 1.1191 1.1191 1.1190 1.1189
    (0.0015) (0.0014) (0.0014) (0.0013) (0.0011)
    (60, 15) 1.1256 1.1255 1.1255 1.1254 1.1253
    (0.0015) (0.0014) (0.0014) (0.0013) (0.0012)
    1.1247 1.1246 1.1246 1.1245 1.1244
    (0.0016) (0.0015) (0.0015) (0.0014) (0.0012)
    1.1195 1.1194 1.1194 1.1193 1.1192
    (0.0016) (0.0015) (0.0015) (0.0013) (0.0011)
    (60, 20) 1.1257 1.1256 1.1256 1.1255 1.1254
    (0.0016) (0.0015) (0.0015) (0.0014) (0.0013)
    1.1248 1.1247 1.1247 1.1246 1.1245
    (0.0017) (0.0016) (0.0016) (0.0015) (0.0013)
    1.1196 1.1195 1.1195 1.1194 1.1193
    (0.0017) (0.0016) (0.0016) (0.0015) (0.0012)

     | Show Table
    DownLoad: CSV
    Table 5.  ACL and CP of 95% CIs of α and β.
    α β
    (n, m) CS MLE MCMC MLE MCMC
    (30, 20) 1.1444 0.0234 0.0751 0.0016
    (0.9435) (0.9446) (0.9332) (0.9418)
    1.1063 0.0236 0.0714 0.0015
    (0.9252) (0.9367) (0.9405) (0.942)
    1.1302 0.0243 0.0688 0.0015
    (0.9572) (0.9253) (0.9396) (0.9549)
    (40, 15) 1.2277 0.0268 0.0781 0.0017
    (0.9698) (0.9507) (0.9377) (0.9456)
    1.3317 0.0283 0.078 0.0016
    (0.9625) (0.947) (0.9608) (0.9655)
    1.301 0.0281 0.0716 0.0015
    (0.951) (0.9606) (0.9712) (0.9447)
    (40, 20) 1.0855 0.0231 0.0707 0.0015
    (0.9615) (0.9566) (0.9573) (0.9539)
    1.0899 0.0232 0.0698 0.0015
    (0.9256) (0.947) (0.9408) (0.959)
    1.077 0.0227 0.0632 0.0013
    (0.9468) (0.9568) (0.9461) (0.9616)
    (50, 15) 1.2419 0.0262 0.0754 0.0016
    (0.9325) (0.9845) (0.9781) (0.9863)
    1.3984 0.0293 0.0776 0.0016
    (0.9124) (0.9683) (0.9786) (0.9452)
    1.2576 0.0274 0.0694 0.0015
    (0.9568) (0.9565) (0.9468) (0.9478)
    (60, 15) 1.2813 0.0269 0.0719 0.0015
    (0.9461) (0.9616) (0.9256) (0.947)
    1.2959 0.0278 0.075 0.0016
    (0.9605) (0.9641) (0.9468) (0.9568)
    1.2552 0.0275 0.0686 0.0015
    (0.9416) (0.9486) (0.9568) (0.9665)
    (60, 20) 1.2832 0.0270 0.0720 0.0016
    (0.9462) (0.9617) (0.9257) (0.948)
    1.2960 0.0279 0.076 0.0017
    (0.9606) (0.9642) (0.94670) (0.9569)
    1.2553 0.0276 0.0687 0.0016
    (0.9417) (0.9487) (0.9569) (0.9666)

     | Show Table
    DownLoad: CSV
    Table 6.  ACL and CP of 95% CIs of S(t) and h(t).
    S(t) h(t)
    (n, m) CS MLE MCMC MLE MCMC
    (30, 20) 0.0755 0.0029 1.2694 0.0022
    (0.9511) (0.9489) (0.952) (0.9472)
    0.10705 0.003 1.2627 0.0021
    (0.9626) (0.9546) (0.9539) (0.9278)
    0.0687 0.003 1.2572 0.0021
    (0.9576) (0.9661) (0.9361) (0.9429)
    (40, 15) 0.0836 0.0033 1.2711 0.0024
    (0.9421) (0.9507) (0.927) (0.9296)
    0.081 0.0034 1.2593 0.0024
    (0.9325) (0.9458) (0.9736) (0.9856)
    0.0748 0.0033 1.2469 0.0022
    (0.958) (0.9603) (0.9601) (0.9602)
    (40, 20) 0.0739 0.0029 1.2693 0.0022
    (0.9462) (0.9467) (0.9502) (0.9506)
    0.07 0.0029 1.2633 0.0021
    (0.9256) (0.9592) (0.9704) (0.9694)
    0.0645 0.00327 1.252 0.0019
    (0.9544) (0.9581) (0.9382) (0.9454)
    (50, 15) 0.0839 0.0032 1.2671 0.0024
    (0.9814) (0.9145) (0.9745) (0.9635)
    0.0832 0.0035 1.2541 0.0023
    (0.9748) (0.9862) (0.9742) (0.9681)
    0.0735 0.0032 1.2483 0.0022
    (0.9254) (0.9438) (0.9152) (0.9634)
    (60, 15) 0.0835 0.0032 1.2625 0.0023
    (0.9637) (0.9504) (0.9667) (0.9567)
    0.0803 0.0033 1.2605 0.0024
    (0.9617) (0.9325) (0.9486) (0.9479)
    0.0735 0.0032 1.2489 0.0021
    (0.9672) (0.9488) (0.9457) (0.9639)
    (60, 20) 0.0836 0.0033 1.2626 0.0024
    (0.9638) (0.9505) (0.9668) (0.9568)
    0.0804 0.0034 1.2606 0.0025
    (0.9618) (0.9326) (0.9487) (0.9480)
    0.0736 0.0033 1.2490 0.0022
    (0.9673) (0.9489) (0.9458) (0.940)

     | Show Table
    DownLoad: CSV

    The following conclusions can be made based on the results:

    (1) According to Tables 14, MSEs decrease as sample size increases, and Bayes estimates get the smallest MSEs for α, β,S(t)and h(t). As a result, in every case that has been studied, Bayes estimates outperform MLE approaches.

    (2) With c=2, the Bayes estimator under LINEX provides superior results in terms of smaller MSEs.

    (3) Scheme Ⅱ is better than schemes Ⅰ and Ⅲ in terms of decreased MSEs for a specified sample n and m effective size values.

    (4) It is evident from Tables 5 and 6 that, for various sample sizes, observed failures, and schemes, CRIs provide more accurate results than ACIs.

    In the following subsections, two real data selected from the medical field are examined. The purpose of this study is to test the suitability of the proposed RTFIW distribution with these data sets.

    The first data set of failure numbers for 20 patients' analgesic relief times (in minutes) is used for the suggested estimation methods. This lifetime data set was introduced by Gross and Clark [26] and is associated with relief times (in minutes) of patients taking an analgesic. These data are provided in Table 7. By comparing the new model RTFIW with the modified Fréchet-exponential distribution (MFED) and exponentiated exponential (EE) under the data set where the value of b is the maximum value of the data. Table 8 shows the MLEs for the distributions. To assess the fitted models, the following measures were used: Bayesian information criterion (BIC), consistent Akaike information criterion (CAIC), Akaike information criterion (AIC), Hannan-Quinn information criterion (HQIC), Kolmogorov-Smirnov (K-S) and the negative log-likelihood (-Log), to seek out the model with the lowest information criterion and good fit parameters. The numerical results in Table 9 show that the RTFIW model fits the data better than other models. The considered data is fitted as shown in Figure 7.

    Table 7.  Data lifetimes of patient-relief times.
    1.1 1.4 1.3 1.7 1.9 1.8 1.6 2.2 1.7 2.7
    1.8 1.5 1.2 1.4 3 1.7 2.3 1.6 2 4.1

     | Show Table
    DownLoad: CSV
    Table 8.  The estimates α and β under progressive type-Ⅱ.
    Model α β θ λ
    MFED 32.433 - 2.0038 -
    EE 36.68 - 2.235
    RTFIW 6.7199 3.7354 - -

     | Show Table
    DownLoad: CSV
    Table 9.  The estimates of the goodness-of-fit for relief times of patients.
    Model KS p-Value -logL AIC BIC HQIC
    MFED 0.120612 0.899975 15.9512 35.9024 37.8938 36.2911
    EE 0.1343 0.817352 16.2606 36.5212 38.5125 36.91
    RTFIW 0.100217 0.955062 14.9239 33.8478 35.8392 35.6581

     | Show Table
    DownLoad: CSV
    Figure 7.  Fitted cdf, fitted pdf, QQ-plots and PP-plots of RTFIWD for data set.

    The empirical failure data distribution's Kolmogorov-Smirnov (K-S) distance from the RTFIW distribution's CDF is 0.100217, with a P-value of 0.955062. As a result, the RTFIW distribution fits the provided data. From n=20 failure data, a progressive type-Ⅱ censored a sample of size m=10 with R=(6,4,0,0,0,0,0,0,0,0) based on the information that these generated. The MLE are computed to be ˆα,ˆβ,ˆS(t) and ˆh(t) such that t=0.4 depend on the progressive type-Ⅱ data reported in Table 7 are shown in Table 10. The Bayes estimator in relation to SEL and LINEX functions are calculated and are also shown in Table 10. From Table 11, the 95% ACIs and CRIs are computed for parameters. Additionally, the 95% ACIs and CRIs for S(t) and h(t) are derived using the delta approach, with the results displayed in Table 11.

    Table 10.  Estimation value for α, β, S(t) and h(t).
    Baysian
    Parameters MLE SLE LINEX
    c = - 2 c = 0.0001 c = 2
    α 1.16548 1.15804 1.15869 1.15804 1.15739
    β 2.85092 2.78356 2.796 2.78356 2.76981
    S 0.999772 0.99954 0.999541 0.99954 0.99954
    h 0.0109181 0.0184367 0.0185916 0.0184367 0.0182849

     | Show Table
    DownLoad: CSV
    Table 11.  95% CIs of α, β, S(t) and h(t).
    Parameter MLE MCMC
    α (0.337871, 1.99309) (1.11849, 1.2089)
    β (0.8467, 4.85513) (2.55267, 2.95076)
    S(0.4) (0.996234, 1.00331) (0.998601, 0.999898)
    h(0.4) (0.131477, 0.153313) (0.00553594, 0.0472466)

     | Show Table
    DownLoad: CSV

    Substituting the MLEs of the unknown parameters α and β for the data set into the inverse of the observed information matrix, we get an estimation of the variance-covariance matrix as follows:

      I1=(3.562041.097231.097230.607173).

    The two sided 95% confidence interval ranges are [1.9207,13.7014] and [2.5453,4.92549], for the parameters α and β respectively.

    The second data set given by Abouammoh et al. [27] represents 40 patients suffering from leukemia in one of the Ministry of Health Hospitals in Saudi Arabia. The survival times of those patients are recorded as follows:

    11518125541844146151673974392498310241062106311651191122212517898078651277129013571357136914081369140814551455147812221251127712901455147812221549157815781599160316051852169617351815

    We aim to compare the new model RTFIW distribution with exponentiated generalized inverse Weibull (EGIW) distribution and generalized inverse Weibull (GIW) distribution under the data set where the value of b is the maximum value of the data.

    Table 12 shows the MLEs for the distributions. To assess the fitted models, the following measures were used: Bayesian information criterion (BIC), Consistent Akaike information criterion (CAIC), Akaike information criterion (AIC), Hannan-Quinn information criterion (HQIC), Kolmogorov-Smirnov (K-S) and the negative log-likelihood (-Log), to seek out the model with the lowest information criterion and good fit parameters. The numerical results in Table 13 show that the RTFIW model fits the data better than other models. The considered data is fitted as shown in Figure 8.

    Table 12.  The estimates of parameters.
    Model α β θ λ
    GIW 0.792 0.298 3.261
    EGIW .933 1.09 0.453 1.102
    RTFIWD 1.10716 1.37948

     | Show Table
    DownLoad: CSV
    Table 13.  The estimates of the goodness-of-fit for the lifetime of 40 patients.
    Model KS PValue -2logL AIC BIC CAIC
    GIW 0.737 0.7584 795.582 801.582 806.648 802.248
    EGIW 0.831 .82464 709.471 717.471 724.226 718.614
    RTFIWD 1.001 0.9744 594.72585 598.7258 602.1036 598.325

     | Show Table
    DownLoad: CSV
    Figure 8.  Fitted cdf, fitted pdf, QQ-plots and PP-plots of RTFIWD for the dataset.

    From n=40 failure data, a progressive type-Ⅱ censored a sample of size m= 20 with R=(10,5,5,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0) based on the information that these generated. The MLEs computed for ˆα,ˆβ,ˆS(t) and ˆh(t) such that t=0.4 and b=1815 based on the progressive type-Ⅱ data are shown in Table 14. The Bayes estimates of SEL and LINEX functions are calculated and are also shown in Table 14.

    Table 14.  Estimation values for α,β,S(t) and h(t) with b=1815.
    Parameters MLE Baysian
    SEL LINEX
    c=2 c=0.0001 c=2
    α 1.10716 1.11602 1.11604 1.11601 1.11602
    β 1.37948 1.3736 1.37364 1.37357 1.3736
    S 0.94189 0.942492 0.942492 0.942491 0.942492
    h 0.490308 0.484726 0.484731 0.48472 0.484726

     | Show Table
    DownLoad: CSV

    From Table 15, the 95% ACIs and CRIs are computed for parameters. Furthermore, the delta method is employed to derive the 95% ACIs and CRIs for S(t) and h(t), hence, the results are shown in Table 15.

    Table 15.  95% CIs of αβS(t) and h(t).
    Parameter MLE MCMC
    α (0.545939,1.66837) (1.1084,1.12323)
    β (0.701971,2.057) (1.36354,1.38614)
    S(0.4) (0.821262,1.06252) (0.941573,0.94353)
    h(0.4) (0.0904752,1.07109) (0.478703,0.489044)

     | Show Table
    DownLoad: CSV

    This study introduces a new truncated distribution called the RTFIW distribution. The new distribution's statistical properties are derived. Using progressive type-Ⅱ censoring, this study aims to provide a number of methods for generating confidence intervals for RTFIW distribution parameters and derivations. The time-relief periods of patients receiving analgesic treatment were studied in addition to Leukemia patient recovery numbers by using various estimation techniques proposed in this study. Different confidence intervals are provided when the MLEs of the unknown parameters are calculated. On the other hand, a simulation study was carried out to assess the accuracy of the suggested estimations. As a result, it is demonstrated that Bayesian approaches perform well in all cases. Due to the pattern of RTFIW's good fit to the applied data, Scheme Ⅱ also performs better than Schemes Ⅰ and Ⅲ in terms of smaller MSEs. In future works, estimation of the RTFIW distribution's parameters using adaptive progressive type-Ⅱ censored data can be obtained, and also can be compared with other types of censored algorithms.

    The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.

    This work was supported by the Deanship of Scientific Research, Vice Presidency for Graduate Studies and Scientific Research, King Faisal University, Saudi Arabia [GRANT No. 5023].

    The authors declare no conflict of interest.



    [1] M. Fréchet, Sur la loi de probabilité de l'écart maximum, Ann. Soc. Math. Polon., 6 (1927), 93–116.
    [2] N. Balakrishnan, Progressive censoring methodology: An appraisal, Test, 16 (2007), 211–259. https://doi.org/10.1007/s11749-007-0061-y doi: 10.1007/s11749-007-0061-y
    [3] G. M. Cordeiro, A. J. Lemonte, On the Marshall-Olkin extended Weibull distribution, Stat. Pap., 54 (2013), 333–353. https://doi.org/10.1007/s00362-012-0431-8 doi: 10.1007/s00362-012-0431-8
    [4] Z. Behdani, G. R. M. Borzadaran, B. Gildeh, Some properties of double truncated distributions and their application in view of income inequality, Comput. Stat., 35 (2020), 359–378. https://doi.org/10.1007/s00180-019-00890-2 doi: 10.1007/s00180-019-00890-2
    [5] L. Zaninetti, A right and left truncated gamma distribution with application to the stars, Adv. Stud. Theor. Phys., 23 (2013), 1139–1147. https://doi.org/10.12988/astp.2013.310125 doi: 10.12988/astp.2013.310125
    [6] M. A. El-Din, A. M. Teamah, A. M. Salem, A. M. T. A. El-Bar, Random sum of mid truncated Lindley distribution, J. Adv. Res. Stat. Probab., 2 (2010), 27–36.
    [7] R. Aggarwala, N. Balakrishnan, Recurrence relations for single and product moments of progressive Type-Ⅱ right censored order statistics from exponential and truncated exponential distributions, Ann. Inst. Stat. Math., 48 (2010), 757–771. https://doi.org/10.1007/BF00052331 doi: 10.1007/BF00052331
    [8] N. Balakrishnan, E. Cramer, The art of progressive censoring, Birkhäuser: NewYork, NY, USA, 2014.
    [9] W. Q. Meeker, L. A. Escobar, F. G. Pascual, Statistical methods for reliability data, John Wiley & Sons: Hoboken, NJ, USA, 2022.
    [10] N. R. Mann, R. E. Schafer, N. D. Singpurwalla, Methods for statistical analysis of reliability and life data, Research Supported by the U. S. Air Force and Rockwell International Corp, New York, John Wiley and Sons, Inc.: Hoboken, NJ, USA, 1974,573.
    [11] J. F. Lawless, Statistical models and methods for lifetime data, Wiley: New York, NY, USA, 1982.
    [12] M. M. Buzaridah, D. A. Ramadan, B. S. El-Desouky, Flexible reduced logarithmic-inverse lomax distribution with application for bladder cancer, Open J. Modell. Simul., 9 (2021), 351–369. https://doi.org/10.4236/ojmsi.2021.94023 doi: 10.4236/ojmsi.2021.94023
    [13] G. C. Montanari, M. Mazzanti, M. Cacciari, J. C. Fothergill, Optimum estimators for the Weibull distribution from censored test data. Progressively censored tests [breakdown statistics], IEEE Trans. Dielectr. Electr. Insul., 5 (2021), 157–164. https://doi.org/10.1109/94.671923 doi: 10.1109/94.671923
    [14] I. Bairamov, S. Eryılmaz, Spacings, exceedances and concomitants in progressive type Ⅱ censoring scheme, J. Stat. Plan. Infer., 136 (2006), 527–536. https://doi.org/10.1016/j.jspi.2004.09.002 doi: 10.1016/j.jspi.2004.09.002
    [15] M. A. W. Mahmoud, D. A. Ramadan, M. M. M. Mansour, Estimation of lifetime parameters of the modified extended exponential distribution with application to a mechanical model, Commun. Stat. Simul. Comput., 1 (2022), 7005–7018. https://doi.org/10.1080/03610918.2020.1821887 doi: 10.1080/03610918.2020.1821887
    [16] D. A. Ramadan, M. Walaa, Estimation of parameters of Alpha power inverse Weibull distribution under progressive type-Ⅱ censoring, J. Stat. Appl. Probab. Lett., 6 (2019), 81–87.
    [17] H. H. Ahmad, E. M. Almetwally, D. A. Ramadan, A comparative inference on reliability estimation for a multi-component stress-strength model under power Lomax distribution with applications, AIMS Math., 7 (2022), 18050–18079. https://doi.org/10.3934/math.2022994 doi: 10.3934/math.2022994
    [18] H. H. Ahmad, K. Elnagar, D. Ramadan, Investigating the lifetime performance index under ishita distribution based on progressive type Ⅱ censored data with applications, Symmetry, 15 (2023), 1779. https://doi.org/10.3390/sym15091779 doi: 10.3390/sym15091779
    [19] M. Shrahili, A. Naif, K. Devendra, A. Salem, Inference for the two-parameter reduced Kies distribution under progressive type-Ⅱ censoring, Mathematics, 8 (2020), 1997. https://doi.org/10.3390/math8111997 doi: 10.3390/math8111997
    [20] C. Luo, L. Shen, A. Xu, Modelling and estimation of system reliability under dynamic operating environments and lifetime ordering constraints, Reliab. Eng. Syst. Saf., 218 (2022), 108–136. https://doi.org/10.1016/j.ress.2021.108136 doi: 10.1016/j.ress.2021.108136
    [21] M. M. Buzaridah, D. A. Ramadan, B. S. El-Desouky, Estimation of some lifetime parameters of flexible reduced logarithmic-inverse lomax distribution under progressive type-Ⅱ censored, Data. J. Math., 2022 (2022). https://doi.org/10.1155/2022/1690458 doi: 10.1155/2022/1690458
    [22] J. J. A. Moors, A quantile alternative for kurtosis, J. Roy. Stat. Soc. D-Sta., 37 (1988), 25–32. https://doi.org/10.2307/2348376 doi: 10.2307/2348376
    [23] R. Aggarwala, N. Balakrishnan, Some properties of progressive censored order statistics from arbitrary and uniform distributions with applications to inference and simulation, J. Stat. Plan. Infer., 70 (1998), 35–49. https://doi.org/10.1016/S0378-3758(97)00173-0 doi: 10.1016/S0378-3758(97)00173-0
    [24] W. Greene, Discrete choice modeling, In: Palgrave Handbook of Econometrics, Palgrave Macmillan: London, UK, 2009,473–556.
    [25] A. Zellner, Bayesian estimation and prediction using asymmetric loss functions, J. Am. Stat. Assoc., 81 (1998), 446–451. https://doi.org/10.1080/01621459.1986.10478289 doi: 10.1080/01621459.1986.10478289
    [26] P. F. Heil, A. J. Gross, V. Clark, Survival distributions: Reliability applications in the biomedical sciences, John Wiley & Sons: Hoboken, NJ, USA, 18 (1975), 501. https://doi.org/10.2307/1268667
    [27] A. M. Abouammoh, S. A. Abdulghani, I. S. Qamber, On partial orderings and testing of new better than renewal used classes, Reliab. Eng. Syst. Safe., 43 (1994), 37–41. https://doi.org/10.1016/0951-8320(94)90094-9 doi: 10.1016/0951-8320(94)90094-9
  • This article has been cited by:

    1. Dina A. Ramadan, Yusra A. Tashkandy, M. E. Bakr, Oluwafemi Samson Balogun, Mustafa M. Hasaballah, Analysis of Marshall–Olkin extended Gumbel type-II distribution under progressive type-II censoring with applications, 2024, 14, 2158-3226, 10.1063/5.0210905
  • Reader Comments
  • © 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1167) PDF downloads(46) Cited by(1)

Figures and Tables

Figures(8)  /  Tables(15)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog