On the identifiability of multivariate survival distribution functions

Let (T₁, T₂) be a non-negative random vector which is subjected to censoring random intervals [X₁, Y₁] and [X₂, Y₂]. The censoring mechanism is such that the available informations on T₁ and T₂ are expressed by a pair of random vectors W=(W₁, W₂) and δ=(δ₁, δ₂), where W_i=max(min(Y_i, T_i), X_i) and In this paper we will show that under some mild conditions the joint survival function of T₁ and T₂ can be expressed uniquely as functional of observable joint survival functions. Our results extend recent works on the randomly right censored bivariate data case and on the univariate problem with double censoring to the bivariate data with double censoring.     

On a new stochastic global optimization algorithm based on censored observations

In this paper a new algorithm is proposed for global optimization problems. The main idea is that of modifying a standard clustering approach by sequentially sampling the objective function while adaptively deciding an appropriate sample size. Theoretical as well as computational results are presented.     

Asymptotic normality of wavelet density estimator under censored dependent observations

In this paper, we discuss the asymptotic normality of the wavelet estimator of the density function based on censored data, when the survival and the censoring times form a stationary ??-mixing sequence. To simulate the distribution of estimator such that it is easy to perform statistical inference for the density function, a random weighted estimator of the density function is also constructed and investigated. Finite sample behavior of the estimator is investigated via simulations too.     

Bayesian Nonparametric Analysis for a Generalized Dirichlet Process Prior

This paper considers a generalization of the Dirichlet process which is obtained by suitably normalizing superposed independent gamma processes having increasing integer-valued scale parameter. A comprehensive treatment of this random probability measure is provided. We prove results concerning its finite-dimensional distributions, moments, predictive distributions and the distribution of its mean. Most expressions are given in terms of multiple hypergeometric functions, thus highlighting the interplay between Bayesian Nonparametrics and special functions. Finally, a suitable simulation algorithm is applied in order to compute quantities of statistical interest.     

Asymptotics of the goodness-of-fit test for a partial linear model with randomly censored data

In this paper, we discuss the problem of testing the hypothesis that the underlying regression is a partial linear model. A test statistic, which is based on the quadratic form of a cusum process of residuals, is proposed. The asymptotic distributions of the test statistic under null hypothesis and the local alternative hypothesis are given. The number simulation shows that the test is available.     

On the marginal distribution of a first order autoregressive process

The question of what marginal distributions are possible for a first order autoregressive process is addressed. Results concerning the possible multimodality of the marginal distribution are obtained.     

Assessing the lifetime performance index of products with the exponential distribution under progressively type II right censored samples

In practice, lifetime performance index C_L is used to measure the potential and performance of a process, where L is the lower specification limit. Progressive censoring scheme is quite useful in many practical situations where budget constraints are in place or there is a demand for rapid testing. In this paper, under the assumption of exponential distribution, this study constructs a maximum likelihood estimator (MLE) of C_L based on the progressively type II right censored sample. The MLE of C_L is then utilized to develop the hypothesis testing procedure in the condition of known L. The new testing procedure can be employed by product managers to assess whether the lifetime of products (or items) adheres to the required level in the condition of known L. Finally, we give one example to illustrate the use of the testing algorithmic procedure under given significance level.     

On a characterization of the exponential distribution by spacings

Summary LetX be a non-negative random variable with probability distribution functionF. SupposeX _i,n (i=1,…,n) is theith smallest order statistics in a random sample of sizen fromF. A necessary and sufficient condition forF to be exponential is given which involves the identical distribution of the random variables (n−i)(X _i+1,n−X_i,n) and (n−j)(X _j+1,n−X_j,n) for somei, j andn, (1≦i<j<n). The work was partly completed when the author was at the Dept. of Statistics, University of Brasilia, Brazil.     

On the inconsistency of Bayesian non-parametric estimators in competing risks/multiple decrement models

In the competing risks/multiple decrement model, the joint distribution is often not identifiable given only the observed time of failure and the cause of failure. The traditional approach is consequently to assume a parametric model. In this paper we shall not do this, but rather assume a Bayesian stance, take a Dirichlet process as a prior distribution, and then calculate the posterior distribution given the data. In this paper we show that in dimensions ? 2, the posterior mean yields an inconsistent estimator of the joint probability law, contrary to the common assumption that the prior law ‘washes out’ with large samples. For single decrement mortality tables however, the non-parametric Bayesian method allows a flexible method for adjusting a standard mortality table to reflect mortality experience, or covariate information.     

On the increments distribution of stock prices

Abstract. In this paper, the models of increment distributions of stock price are constructed with two approaches. The first approach is based on limit theorems of random summation. The second approach is based on the statistical analysis of the increment distribution of the logarithms of stock prices.     

�ݺ���ģ���º����������ķDz�����Ҷ˹����

??The linear accelerated model is often used to the statistical analysis of constant stress accelerated life test, whereas it does not relate well with the facts. By adopting the power functional accelerated model, the relationship of sample quantiles among different constant stress levels is obtained, which can lead to the estimations of the parameters in accelerated model and the characteristic coefficient vectors by virtue of the least square method, then the life-time data transformation between different stress levels can be operated. For complete data and censoring data, a Dirichlet process prior is introduced to gain the posterior distribution and the nonparametric Bayesian estimation of the reliability function, meanwhile, the consistency of the posterior estimators is proved. Finally, a real life example of Metal-Oxide-Semiconductor capacitors is analyzed to illustrate the effect of our model.     

On the survival probability of generalized nearest-particle systems

A class of spatial growth models, including reversible nearest-particle systems as special cases, is defined via the range of an underlying random walk. Lower and upper bounds for the survival probability are established under various assumptions. The proofs rely on potential theoretic results for the random walk.     

On the transition from a Markov chain to a continuous time process

Starting from a real-valued Markov chain X₀,X₁,…,X_n with stationary transition probabilities, a random element {Y(t);t[0, 1]} of the function space D[0, 1] is constructed by letting Y(k/n)=X_k, k= 0,1,…,n, and assuming Y (t) constant in between. Sample tightness criteria for sequences {Y(t);t[0,1]};_n of such random elements in D[0, 1] are then given in terms of the one-step transition probabilities of the underlying Markov chains. Applications are made to Galton-Watson branching processes.     

On the limiting behavior of randomly weighted partial sums

We study the almost sure limiting behavior and convergence in probability of weighted partial sums of the form where {W_nj, 1jn, n1} and {X_nj, 1jn, n1} are triangular arrays of random variables. The results obtain irrespective of the joint distributions of the random variables within each array. Applications concerning the Efron bootstrap and queueing theory are discussed.     

A representation for the posterior distribution of the location of a moving target

A target moves according to a continuous stochastic process in Euclidean R_^H. A search is conducted by choosing a search strategy and by observing the state of a detection process. Methods for representing the posterior distributions for target location given no detection are discussed. In particular, methods for parameterizing Gaussian models for target motion and the transition intensity of the detection process which yield tractable representations are introduced. The methodology is discussed in terms of two examples.     

On the output process of an M/M/1 queue with randomly varying system parameters

In this paper an analysis of the output process from an M/M/1 queue where the arrival and service rates vary randomly is presented. The results include expressions for the mean, variance and distribution of the interdeparture interval, the joint density function of two successive interdeparture intervals and their correlation. An interesting feature of the results is that the moments of the interdeparture time are expressed in terms of the expected times to first and second departures from an arbitrary point in time.     

Bootstrapping the empirical distribution of a linear process

The validity of the moving block bootstrap for the empirical distribution of a short memory causal linear process is established under simple conditions that do not involve mixing or association. Sufficient conditions can be expressed in terms of the existence of moments of the innovations and summability of the coefficients of the linear model. Applications to one and two sample tests are discussed.     

On the loss of information due to fuzziness in experimental observations

The absence of exactness in the observation of the outcomes of a random experiment always entails a loss of information about the experimental distribution. This intuitive assertion will be formally proved in this paper by using a mathematical model involving the notions of fuzzy information and fuzzy information system (as intended by Tanaka, Okuda and Asai) and Zadeh's probabilistic definition. On the basis of this model we are first going to consider a family of measures of information enclosing some well-known measures, such as those defined by Kagan, Kullback-Leibler and Matusita, and then to establish methods for removing the loss of information due to fuzziness by increasing suitably the number of experimental observations.