• Journal of the Korean Statistical Society
• /
• 제30권3호
• /
• pp.481-498
• /
• 2001

Empirical Bayes procedure of nonparametric estiamtion of cumulative hazard rates based on censored data is considered using the beta process priors of Hjort(1990). Beta process priors with unknown parameters are used for cumulative hazard rates. Empirical Bayes estimators are suggested and asymptotic optimality is proved. Our result generalizes that of Susarla and Van Ryzin(1978) in the sensor that (i) the cumulative hazard rate induced by a Dirichlet process is a beta process, (ii) our empirical Bayes estimator does not depend on the censoring distribution while that of Susarla and Van Ryzin(1978) does, (iii) a class of estimators of the hyperprameters is suggested in the prior distribution which is assumed known in advance in Susarla and Van Ryzin(1978). This extension makes the proposed empirical Bayes procedure more applicable to real dta sets.

• International Journal of Reliability and Applications
• /
• 제14권1호
• /
• pp.1-9
• /
• 2013

A nonparametric procedure for testing exponentially against used better than aged in expectation (UBAE) class of life distributions is presented. We construct a test statistics based on scaled total time on test (TTT)-transformation, to test exponentiality against UBAE class of life distributions. The distribution of the statistic is investigated via simulation. Practical applications of the proposed test are presented.

• 한국수학교육학회지시리즈B:순수및응용수학
• /
• 제5권2호
• /
• pp.123-132
• /
• 1998

Let F and G denote the distribution functions of the failure times and the censoring variables in a random censorship model. Susarla and Van Ryzin(1978) verified consistency of $F_{\alpha}$, he NPBE of F with respect to the Dirichlet process prior D($\alpha$), in which they assumed F and G are continuous. Assuming that A, the cumulative hazard function, is distributed according to a beta process with parameters c, $\alpha$, Hjort(1990) obtained the Bayes estimator $A_{c,\alpha}$ of A under a squared error loss function. By the theory of product-integral developed by Gill and Johansen(1990), the Bayes estimator $F_{c,\alpha}$ is recovered from $A_{c,\alpha}$. Continuity assumption on F and G is removed in our proof of the consistency of $A_{c,\alpha}$ and $F_{c,\alpha}$. Our result extends Susarla and Van Ryzin(1978) since a particular transform of a beta process is a Dirichlet process and the class of beta processes forms a much larger class than the class of Dirichlet processes.

• International Journal of Reliability and Applications
• /
• 제10권2호
• /
• pp.81-88
• /
• 2009

A non-parametric procedure is presented for testing exponentially against used better than aged in convex ordering class (UBAC) of life distributions based on u-test. Convergence of the proposed statistic to the normal distribution is proved. Selected critical values are tabulated for sample sizes 5(5)40. The Pitman asymptotic relative efficiency of my proposed test to tests of other classes is studied. An example of 40 patients suffering from blood cancer disease demonstrates practical application of the proposed test.

• Journal of the Korean Data and Information Science Society
• /
• 제7권1호
• /
• pp.59-68
• /
• 1996

In this paper, we consider how to estimate New Better Than Used (NBU) survival curves from randomly right censored data. We propose several possible NBU estimators and study their properties. Numerical studies indicate that the proposed estimators are appropriate in practical use. Some useful examples are presented.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(3)
• /
• pp.9-12
• /
• 2001

Learning vector quantization (LVQ) is a supervised learning technique that uses class information to move the vector quantizer slightly, so as to improve the quality of the classifier decision regions. In this paper we propose a selection method of initial codebook vectors for a teaming vector quantization (LVQ3) using support vector machines (SVM). The method is experimented with artificial and real design data sets and compared with conventional methods of the condensed nearest neighbor (CNN) and its modifications (mCNN). From the experiments, it is discovered that the proposed method produces higher performance than the conventional ones and then it could be used efficiently for designing nonparametric classifiers.

• 한국통신학회논문지
• /
• 제28권4C호
• /
• pp.374-378
• /
• 2003

이 논문에서는 신호의존성 잡음과 가산성 잡음이 함께 있을 때 확률 신호를 찾아내는 두 표본을 쓰는 국소 최적 순위 검파기의 점근 성능을 다룬다. 비모수 검파기인 국소 최적 순위 검파기는 국소 최적 검파기와 비교하여 볼 때, 시간상으로 상관이 있는 확률 신호에 대해서 거의 같은 점근 성능을 가진다. 두 표본을 쓰는 국소 최적 순위 검파기는 한 표본을 쓰는 국소 최적 순위 검파기와 같게 수행한다는 점은 주목할 만하다.

• Journal of the Korean Data and Information Science Society
• /
• 제2권
• /
• pp.45-62
• /
• 1991

신뢰수명의 검정이나 임상실험에서 대상물에 대한 관측치를 충분히 얻기 위해서는 많은 시간과 경비가 필요하거나 현실적인 제한으로 인하여 관측이 불가능한 경우가 흔히 있다. 이러한 이유로 점진적 임의중단법에 의하여 얻어진 관측치를 이용한 생존함수의 추정은 현실적으로 매우 중요하다고 하겠다. 연구에서는 점진적 임의중단된 자료를 기초로 스플라인함수를 이용하여 생존함수의 비모수적 최우추정량을 제안하고 그것의 성질과 효율성을 비교 연구한다.

• Journal of the Korean Statistical Society
• /
• 제30권1호
• /
• pp.127-138
• /
• 2001

In order to combine parametric and nonparametric approaches together for survival analysis with censored observations, a new class of priors called mixtures of the beta processes is introduced. It is shown that mixtures of beta processes priors generalized the well known priors - mixtures of Dirichlet processes, and they are conjugate with right censored observations. Formulas for computing the posterior distribution are derived. Finally, a real data set is analyzed for illustrational purpose.

• International Journal of Reliability and Applications
• /
• 제3권2호
• /
• pp.91-98
• /
• 2002

A class of life distributions, whose mean residual life keeps stable at its earlier phase and then starts to decrease in time, is proposed to model the life of an element haying survived its burn-in. A strongly consistent estimator and a nonparametric testing procedure are developed to locate the occurrence of the change-point of the mean residual life. Finally, some numerical simulations are employed to be an illustration as well.