Testing behavior of the reversed hazard rate

The concept of reversed hazard rate of a random life is defined as the ratio between the life probability density to its distribution function. This concept plays a role in analyzing censored data and is applicable in such areas as Forensic Sciences. In this investigation, we address the question of testing the reversed hazard rate where the null is that the reversed hazard rate is an assigned function while the alternative is that it is decreasing but not equal to the null function. Two approaches are discussed: one is based on the empirical distribution function while the other is based on the celebrated kernel methods. The limiting distributions of the test statistics are given and its asymptotic Pittman efficiencies are evaluated for well-known alternatives when the null distribution is exponential. Some other related problems are also addressed.     

Inequalities for order statistics of samples from distributions with monotone hazard rate

A number of inequalities relating the mathematical expectations of order statistics or functions of order statistics in the case where the sample is taken from a distribution with monotone hazard rate function are presented. These inequalities can be applied in reliability theory and other areas of probability theory and mathematical statistics. In particular, in the paper, they are used to obtain new characterizations of the exponential distribution.     

Additive–multiplicative hazards model with current status data

The additive–multiplicative hazards (AMH) regression model specifies an additive and multiplicative form on the hazard function for the counting process associated with a multidimensional covariate process, which contains the Cox proportional hazards model and the additive hazards model as its special cases. In this paper, we study the AMH model with current status data, where the cumulative hazard hazard function is assumed to be nonparametric and is estimated using B-splines with monotonicity constraint on the functional, while a simultaneous sieve maximum likelihood estimation is proposed to estimate regression parameters. The proposed estimator for the parameter vector is shown to be asymptotically normal and semiparametric efficient. The B-splines estimator of the functional of the cumulative hazard function is shown to achieve the optimal nonparametric rate of convergence. A simulation study is conducted to examine the finite sample performance of the proposed estimators and algorithm, and a real data example is presented for illustration.     

Generalized mixtures in reliability modelling: Applications to the construction of bathtub shaped hazard models and the study of systems

In this paper, we obtain and discuss some general properties of hazard rate (HR) functions constructed via generalized mixtures of two members. These results are applied to determine the shape of generalized mixtures of an increasing hazard rate (IHR) model and an exponential model. In addition, we note that these kind of generalized mixtures can be used to construct bathtub‐shaped HR models. As examples, we study in detail two cases: when the IHR model chosen is a linear HR function and when the IHR model is the extended exponential‐geometric distribution. Finally, we apply the results and show the utility of generalized mixtures in determining the shape of the HR function of different systems, such as mixed systems or consecutive k‐out‐of‐n systems. Copyright © 2008 John Wiley & Sons, Ltd.     

A new method for proving weak convergence results applied to nonparametric estimators in survival analysis

Using the limit theorem for stochastic integral obtained by Jakubowski et al. (Probab. Theory Related Fields 81 (1989) 111–137), we introduce in this paper a new method for proving weak convergence results of empirical processes by a martingale method which allows discontinuities for the underlying distribution. This is applied to Nelson–Aalen and Kaplan–Meier processes. We also prove that the same conclusion can be drawn for Hjort's nonparametric Bayes estimators of the cumulative distribution function and cumulative hazard rate.     

The Log–Lindley distribution as an alternative to the beta regression model with applications in insurance

In this paper a new probability density function with bounded domain is presented. The new distribution arises from the generalized Lindley distribution proposed by Zakerzadeh and Dolati (2010). This new distribution that depends on two parameters can be considered as an alternative to the classical beta distribution. It presents the advantage of not including any special function in its formulation. After studying its most important properties, some useful results regarding insurance and inventory management applications are obtained. In particular, in insurance, we suggest a special class of distorted premium principles based on this distribution and we compare it with the well-known power dual premium principle. Since the mean of the new distribution can be normalized to give a simple parameter, this new model is appropriate to be used as a regression model when the response is bounded, being therefore an alternative to the beta regression model recently proposed in the statistical literature.     

Diffusion and survival models for the process of entry into marriage

In this paper different survival models with a non‐parametric hazard rate function are applied to the process of entry into marriage. The hazard function of the Hernes model of the marriage process as well as the log‐logistic survival model are both derivable from a differential equation for a social diffusion process. The log‐logistic model might be appropriate for modelling the marriage rate because of its non‐monotonic hazard function. Moreover, in light of our analysis, application of this model to the marriage process can be justified by the theoretical rationale of a process of social diffusion. Both models are tested using German age‐at‐marriage data and the U.S. cohort data analyzed by Hernes (1972). It can be shown that the three parameter Hernes model yields a good fit while the log‐logistic model, with only two parameters, leads to a moderate approximation of the data.     

A Berry-Esséen Bound for the Product-Limit Estimator under Left-Truncation and Right-Censoring

For left-truncated and right-censored data, the product-limit estimator F?_xis a well-known nonparametric estimator for the distribution function F_x of the target variable X such as the survival time. Since F?_xas a very complicated product form we establish first the Berry-Esseen bound for the cumulative hazard estimator of F_x The cumulative hazard estimator can be represented as a U-statistic. By using the result in Helmers and van Zwet [6], we derive the Berry-esséen bound for this U-statistic. Then Berry-Esseen bounds for the distribution of the cumulative hazard estimator and the normal distribution and the distribution of the product-limit estimator and the normal distribution are obtained.     

The new Neyman type A beta Weibull model with long-term survivors

For the first time, we propose a flexible cure rate survival model by assuming that the number of competing causes of the event of interest follows the Neyman type A distribution and the time to this event has the beta Weibull distribution. This new model can be used to analyze survival data when the hazard rate function is increasing, decreasing, bathtub or unimodal-shaped. It includes some commonly used lifetime distributions and some well-known cure rate models as special cases. Maximum likelihood and non-parametric bootstrap are used to estimate the regression parameters. We derive the appropriate matrices for assessing local influence on the parameter estimates under different perturbation schemes and present some ways to perform global influence analysis. The usefulness of the new model is illustrated by means of an application in the medical area.     

Some beta distributions

Four new generalizations of the standard beta distribution are introduced. Various properties are derived for each distribution, including its hazard rate function and moments.     

A new generalized linear exponential distribution and its applications

A new generalized linear exponential distribution （NCLED） is considered in this paper which can be deemed as a new and more flexible extension of linear exponential distribution. Some statistical properties for the NGLED such as the hazard rate function, moments, quantiles are given. The maximum likelihood estimations （MLE） of unknown parameters are also discussed. A simulation study and two real data analyzes are carried out to illustrate that the new distribution is more flexible and effective than other popular distributions in modeling lifetime data.     

The monotonicity of the reliability measures of the beta distribution

In this paper, we investigate the monotonic properties of the hazard (failure) rate and mean residual life function (life expectancy) of the beta distribution. The monotonic properties are sometimes very useful in identifying an appropriate model.     

Reliability function of a class of time-dependent systems with standby redundancy

By applying shortest path analysis in stochastic networks, we introduce a new approach to obtain the reliability function of time-dependent systems with standby redundancy. We assume that not all elements of the system are set to function from the beginning. Upon the failure of each element of the active path in the reliability graph, the system switches to the next path. Then, the corresponding elements are activated and consequently the connection between the input and the output is established. It is also assumed each element exhibits a constant hazard rate and its lifetime is a random variable with exponential distribution. To evaluate the system reliability, we construct a directed stochastic network called E-network, in which each path corresponds with a minimal cut of the reliability graph. We also prove that the system failure function is equal to the density function of the shortest path of E-network. The shortest path distribution of this new constructed network is determined analytically using continuous-time Markov processes.     

Reliability and non-reliability studies of Poisson variables in series and parallel systems

In this paper, we have derived the distribution of the minimum and maximum of two independent Poisson random variables. A useful procedure for computing the probabilities is given and a total of four numerical examples are presented. Of these four examples, the first two are on the generated data and the other two are on the Champion League Soccer data in order to illustrate the model which is considered here. The hazard rate and the reversed hazard rate, of the minimum and maximum of two independent discrete random variables, are also obtained and their monotonicity is investigated. The results for the Poisson-distributed variables are obtained as special cases.     

The Hazard Rate and the Reversed Hazard Rate Orders, with Applications to Order Statistics

In this paper we first point out a simple observation that can be used successfully in order to translate results about the hazard rate order into results about the reversed hazard rate order. Using it, we derive some interesting new results which compare order statistics in the hazard and in the reversed hazard rate orders; as well as in the usual stochastic order. We also simplify proofs of some known results involving the reversed hazard rate order. Finally, a few further applications of the observation are given.     

Limit of hazard rate function of coherent system with discrete life

In this paper we study the limiting behavior of the hazard rate function of a coherent system 𝒮 whose components are independent and have discrete lifetimes. The lifetime of the system 𝒮 can be viewed as the lifetime of a parallel system consisting of the series systems obtained from the minimal paths of 𝒮. If we appropriately interpret the ‘strongest’ minimal path(s) of the system 𝒮, it will be shown that the limit of the hazard rate functions of 𝒮 is the same as that of the ‘strongest’ path(s) of 𝒮. The application of the main results are also provided. Copyright © 2010 John Wiley & Sons, Ltd.     

�����������������ƽ��ʣ�������Ƚ�

??The mean residual life (MRL) function plays a very important role in the area of reliability engineering, survival analysis, and many other fields. In this paper, we introduce and study a new stochastic order which gives stochastic comparison for mean residual life of strictly increasing concave function of two random variables. We show that this new stochastic order lies between the hazard rate and mean residual life orders. The preservation properties under mixtures are presented here. Finally, we give some applications of this new order in reliability theory.     

Optimizing Preventive Maintenance Models

We deal with the problem of scheduling preventive maintenance (PM) for a system so that, over its operating life, we minimize a performance function which reflects repair and replacement costs as well as the costs of the PM itself. It is assumed that a hazard rate model is known which predicts the frequency of system failure as a function of age. It is also assumed that each PM produces a step reduction in the effective age of the system. We consider some variations and extensions of a PM scheduling approach proposed by Lin et al. [6]. In particular we consider numerical algorithms which may be more appropriate for hazard rate models which are less simple than those used in [6] and we introduce some constraints into the problem in order to avoid the possibility of spurious solutions. We also discuss the use of automatic differentiation (AD) as a convenient tool for computing the gradients and Hessians that are needed by numerical optimization methods. The main contribution of the paper is a new problem formulation which allows the optimal number of occurrences of PM to be determined along with their optimal timings. This formulation involves the global minimization of a non-smooth performance function. In our numerical tests this is done via the algorithm DIRECT proposed by Jones et al. [19]. We show results for a number of examples, involving different hazard rate models, to give an indication of how PM schedules can vary in response to changes in relative costs of maintenance, repair and replacement. Part of this work was carried out while the first author was a Visiting Professor in the Department of Mechanical Engineering at the University of Alberta in December 2003.     

Analyzing right-censored length-biased data with additive hazards model

Length-biased data are often encountered in observational studies, when the survival times are left-truncated and right-censored and the truncation times follow a uniform distribution. In this article, we propose to analyze such data with the additive hazards model, which specifies that the hazard function is the sum of an arbitrary baseline hazard function and a regression function of covariates. We develop estimating equation approaches to estimate the regression parameters. The resultant estimators are shown to be consistent and asymptotically normal. Some simulation studies and a real data example are used to evaluate the finite sample properties of the proposed estimators.