Representing the mean residual life in terms of the failure rate

In survival or reliability studies, the mean residual life or life expectancy is an important characteristic of the model. Whereas the failure rate can be expressed quite simply in terms of the mean residual life and its derivative, the inverse problem—namely that of expressing the mean residual life in terms of the failure rate—typically involves an integral of a complicated expression. In this paper, we obtain simple expressions for the mean residual life in terms of the failure rate for certain classes of distributions which subsume many of the standard cases. Several results in the literature can be obtained using our approach. Additionally, we develop an expansion for the mean residual life in terms of Gaussian probability functions for a broad class of ultimately increasing failure rate distributions. Some examples are provided to illustrate the procedure.     

Some Necessary and Sufficient Conditions for Age Replacement with Non-Zero Downtimes

In this paper we consider an age replacement strategy, where downtimes are non-zero. Although this model is well known, the literature gives no necessary and sufficient conditions for age replacement to be preferred to replacement on failure. In this paper we derive such conditions in terms of the minimum of the mean residual life function. When age replacement is indicated, we derive sufficient conditions for the existence of a global minimum to the asymptotic expected cost rate function. These results are illustrated for the Weibull and Gamma distributions.     

Optimum Burn-in Time for a Bathtub Shaped Failure Distribution

An important problem in reliability is to define and estimate the optimal burn-in time. For bathtub shaped failure-rate lifetime distributions, the optimal burn-in time is frequently defined as the point where the corresponding mean residual life function achieves its maximum. For this point, we construct an empirical estimator and develop the corresponding statistical inferential theory. Theoretical results are accompanied with simulation studies and applications to real data. Furthermore, we develop a statistical inferential theory for the difference between the minimum point of the corresponding failure rate function and the aforementioned maximum point of the mean residual life function. The difference measures the length of the time interval after the optimal burn-in time during which the failure rate function continues to decrease and thus the burn-in process can be stopped.      

On the shape of the mean residual lifetime function

Some general properties of the mean residual life (MRL) function are studied. The analysis is based on the shape of the corresponding failure rate. The conditions under which the failure rate and the reciprocal to the MRL function have asymptotically equivalent behaviour as t→∞ are discussed. The simplest non‐monotone shapes of the functions under consideration (bathtub and upside down bathtub) are also considered. The MRL functions for mixtures of distributions are described via the corresponding conditional probability density functions. The direct proportional model of mixing is characterized and some asymptotic results on the shape of the mixture MRL are obtained. Some simple examples are given. Copyright © 2002 John Wiley & Sons, Ltd.     

负相协平均剩余寿命函数和有效函数的一类非参数估计

本文在 NA 样本下,讨论了平均剩余寿命函数和有效函数的非参数递归型估计的相合性和渐近正态性．     

An asymptotic expansion for the normalizing constant of the Conway–Maxwell–Poisson distribution

The Conway–Maxwell–Poisson distribution is a two-parameter generalization of the Poisson distribution that can be used to model data that are under- or over-dispersed relative to the Poisson distribution. The normalizing constant \(Z(\lambda ,\nu )\) is given by an infinite series that in general has no closed form, although several papers have derived approximations for this sum. In this work, we start by using probabilistic argument to obtain the leading term in the asymptotic expansion of \(Z(\lambda ,\nu )\) in the limit \(\lambda \rightarrow \infty \) that holds for all \(\nu >0\). We then use an integral representation to obtain the entire asymptotic series and give explicit formulas for the first eight coefficients. We apply this asymptotic series to obtain approximations for the mean, variance, cumulants, skewness, excess kurtosis and raw moments of CMP random variables. Numerical results confirm that these correction terms yield more accurate estimates than those obtained using just the leading-order term.

     

Central limit asymptotics for shifts of finite type

We study the rate of convergence and asymptotic expansions in the central limit theorem for the class of Hölder continuous functions on a shift of finite type endowed with a stationary equilibrium state. It is shown that the rate of convergence in the theorem isO(n ^?1/2) and when the function defines a non-lattice distribution an asymptotic expansion to the order ofo(n ^?1/2) is given. Higher-order expansions can be obtained for a subclass of functions. We also make a remark on the central limit theorem for (closed) orbital measures.     

The differential mean value of divided differences

The paper deals with the asymptotic behaviour of the differential mean value of divided differences as the interval shrinks to zero by presenting an asymptotic expansion. The coefficients are given by a recurrence formula. For a wide class of analytic functions the differential mean value can be represented by a convergent sum. Our results generalize two recent theorems by Powers, Riedel and Sahoo [R.C. Powers, T. Riedel, P.K. Sahoo, Limit properties of differential mean values, J. Math. Anal. Appl. 227 (1998) 216-226].     

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.     

High-precision computation of the confluent hypergeometric functions via Franklin-Friedman expansion

We present a method of high-precision computation of the confluent hypergeometric functions using an effective computational approach of what we termed Franklin-Friedman expansions. These expansions are convergent under mild conditions of the involved amplitude function and for some interesting cases the coefficients can be rapidly computed, thus providing a viable alternative to the conventional dichotomy between series expansion and asymptotic expansion. The present method has been extensively tested in different regimes of the parameters and compared with recently investigated convergent and uniform asymptotic expansions.     

Weighted stationary phase of higher orders

The subject matter of this paper is an integral with exponential oscillation of phase f(x) weighted by g(x) on a finite interval [α β]: When the phase f(x) has a single stationary point in (α β), an nth-order asymptotic expansion of this integral is proved for n ≥ 2: This asymptotic expansion sharpens the classical result for n = 1 by M. N. Huxley. A similar asymptotic expansion was proved by V. Blomer, R. Khan and M. Young under the assumptions that f(x) and g(x) are smooth and g(x) is compactly supported on R: In the present paper, however, these functions are only assumed to be continuously differentiable on [α β] 2n + 3 and 2n + 1 times, respectively. Because there are no requirements on the vanishing of g(x) and its derivatives at the endpoints α and β, the present asymptotic expansion contains explicit boundary terms in the main and error terms. The asymptotic expansion in this paper is thus applicable to a wider class of problems in analysis, analytic number theory, and other fields.     

Quantile Function Expansion Using Regularly Varying Functions

We present a simple result that allows us to evaluate the asymptotic order of the remainder of a partial asymptotic expansion of the quantile function h(u) as u → 0⁺ or 1^?. This is focussed on important univariate distributions when h(?) has no simple closed form, with a view to assessing asymptotic rate of decay to zero of tail dependence in the context of bivariate copulas. Motivation of this study is illustrated by the asymptotic behaviour of the tail dependence of Normal copula. The Normal, Skew-Normal and Gamma are used as initial examples. Finally, we discuss approximation to the lower quantile of the Variance-Gamma and Skew-Slash distributions.     

Optimal Robust Design of Step Stress Accelerated Life Test Scheme under Weibull Distribution

The traditional accelerated life test scheme is necessary to give the rough values of some model parameters in advance, but the influence of fluctuation on the stability of test scheme is irregulared. Based on the prior life test information, this paper aims to minimize the mean and variance of asymptotic variance of $p$-quantile life estimate under normal test stress level, using maximum likelihood estimation theory and Nelson cumulative failure principle, the optimal robust design mathematical model of step stress accelerated life test scheme with uncertainty parameters under Weibull distribution is established. The results of optimal robust design of step stress accelerated life test scheme for electrical connectors show that: comparing with the optimal design of step stress test scheme in the literature, the optimal robust design scheme is not sensitive to the uncertainty of model parameters when the asymptotic variance of median life estimate is basically the same; Comparing with the optimal design of constant accelerated life test scheme, when the statistical accuracy of test data is basically the same, the number of samples required can be reduced by 1/5, and the test time can be reduced by about 1/4.     

The extreme points of the set of decreasing failure rate distributions

Summary Since the class of extended decreasing failure rate (EDFR) life distributions (i.e., distributions with support in [0, ]) is compact and convex, it follows from Choquet's Theorem that every EDFR life distribution can be represented as a mixture of extreme points of the EDFR class. We identify the extreme points of this class and of the standard class of decresing failure rate (DFR) life distributions. Further, we show that even though the convex class of DFR life distributions is not compact, every DFR life distribution can be represented as a mixture of extreme points of the DFR class.Research sponsored by the Air Force Office of Scientific Research, AFSC, USAF, under Grant AFOSR 78-3678.Research sponsored by the National Science Foundation MCS-7904698.     

Reliability Studies of Bivariate Distributions with Pareto Conditionals

In this paper we study Arnold's (1987, Statist. Probab. Lett.5, 263–266) class of bivariate distributions with Pareto conditionals from a reliability point of view. Failure rates and mean residual life function of the marginal distributions and their monotonic properties are studied. The hazard components and their properties are investigated and their relationships with some measures of dependence are established. Finally, the failure rate of the minimum of the two components is examined and its monotonicity is investigated. Some of the results presented here are general and would be useful in studying the dependence structure in other classes of bivariate distributions.     

Nonlinear wavelet density estimation with censored dependent data

In this paper, we provide an asymptotic expansion for the mean integrated squared error (MISE) of nonlinear wavelet estimator of survival density for a censorship model when the data exhibit some kind of dependence. It is assumed that the observations form a stationary and α‐mixing sequence. This asymptotic MISE expansion, when the density is only piecewise smooth, is same. However, for the kernel estimators, the MISE expansion fails if the additional smoothness assumption is absent. Also, we establish the asymptotic normality of the nonlinear wavelet estimator. Copyright © 2011 John Wiley & Sons, Ltd.     

基于长偏置数据的一个光滑剩余生命函数估计量(英文)

受希勒引理的启发,针对长偏置数据给出一个带泊松权重的平均剩余生命函数的光滑估计量.同时研究了该光滑估计量的渐进性质,例如强相合性及渐进正态性等性质.     

An extended CEV model and the Legendre transform-dual-asymptotic solutions for annuity contracts

This paper develops an extended constant elasticity of variance (E-CEV) model to overcome the shortcomings of the general CEV model. Under the E-CEV model, we study the optimal investment strategy before and after retirement in a defined contribution pension plan where benefits are paid by annuity. By applying the Legendre transform, dual theory and an asymptotic expansion approach, we respectively derive two asymptotic strategies for a CRRA and CARA utility functions in two different periods. Furthermore, we find that each asymptotic strategy can be decomposed into an optimal zero-order strategy and a perturbation strategy. The optimal zero-order strategy denotes an investment strategy where the current volatility is just equal to the mean level of the volatility, whereas the perturbation strategy provides an approximation solution to hedge the slow varying nature of the current volatility deviating from mean level. Finally, we find that the optimal zero-order strategy under given conditions will reduce to the results of Devolder et al. (2003), Xiao et al. (2007) and Gao (2009), respectively.     

The problem of differentiating an asymptotic expansion in real powers. Part I: Unsatisfactory or partial results by classical approaches

In two papers, the problem of formal differentiation of an asymptotic expansion in the real domain of type $$ f(x) - a_1 x^{\alpha _1 } + \cdots + a_n x^{\alpha _n } + o(x^{\alpha _n } ),x \to + \infty , $$ is amply studied. In Part I, we show that the classical viewpoints and techniques concerning formal differentiation of an asymptotic relation $$ f(x) - ax^\alpha + o(x^\alpha ),x \to + \infty , $$ give either unsatisfactory or partial results when applied to an asymptotic expansion with at least two meaningful terms. Simple examples show that some of these results are the best possible in the classical context. Hence a change of viewpoint is necessary to arrive at useful results.