• 품질경영학회지
• /
• 제23권4호
• /
• pp.157-166
• /
• 1995

Process capability indices are used to determine whether a production process is capable of producing items within a specified tolerance. We could estimate the finite sample distributions of some process capability indices with bootstrap method. In this paper, we derive the asymptotic bootstrap distributions of some process capability indices ${\hat{C}}^*_p$, ${\hat{C}}^*_{pk}$ and ${\hat{C}}^*_{pm}$ under general proper conditions. These asymptotic distributiops would be used in constructing some bootstrap confidence intervals. Also, we examine some small sample properties related to these estimators by some simulations.

• 품질경영학회지
• /
• 제21권2호
• /
• pp.162-169
• /
• 1993

For the problem of testing the parallelism of several regression lines against ordered alternatives, two test statistics and proposed and examined. The proposed statistics are linear combinations of robust estimators of slope parameters, which are modifications of the Adichie (1976) test based on scores. The asymptotic null variances of the proposed states tics are estimated by the kernel density estimation methods. The proposed tests are compared with the Adichie's test in terms of asymptotic relative efficiency and small-sample powers.

• 품질경영학회지
• /
• 제23권4호
• /
• pp.42-51
• /
• 1995

The bootstrap confidence intervals are a computer-based method for assigning measures of accuracy to statistical estimators. In this paper we examine the small sample behavior of the Kaplan-Meier and Nelson-type estimators for the survival function using the bootstrap and asymptotic normal-theory confidence intervals. The Nelson-type estimator is nearly always better than the Kaplan-Meier estimator in the sense of achieved error rates. From the point of confidence length, the reverse is true. Also, we show that the bootstrap confidence intervals are better than the asymptotic normal-theory confidence intervals in terms of achieved error rates and confidence length.

• 한국정밀공학회지
• /
• 제17권12호
• /
• pp.121-128
• /
• 2000

The scope of this paper is focused to the systems which have the time period and they should be necessarily studied in the sense of stability and design method of controller to stabilize the orignal unstable systems. In general, the time periodic systems or the systems having same motions during certain time interval are easily found in rotating motion device, i.e., satellite or helicopter and widely used in factory automation systems. The characteristics of the selected dynamic systems are analyzed with the new stability concept and stabilization control method based on Lyapunov direct method. The new method from Lyapunov stability criteria which satisfies the energy convergence is studied with linear algebraic method. And the numerical procedures are developed with computational programming method to apply to the practical linear periodic systems. The results from this paper demonstrate the usefulness in analysis of the asymptotic stability and stabilization of the unstable linear periodic system by using the developed simulation procedures.

• Communications for Statistical Applications and Methods
• /
• 제12권3호
• /
• pp.557-587
• /
• 2005

This paper elucidates the limiting Gaussian distribution of a class of rank order statistics {$T_N$} for two sample problem pertaining to empirical processes of the squared residuals from two independent samples of GARCH processes. A distinctive feature is that, unlike the residuals of ARMA processes, the asymptotics of {$T_N$} depend on those of GARCH volatility estimators. Based on the asymptotics of {$T_N$}, we empirically assess the relative asymptotic efficiency and effect of the GARCH specification for some GARCH residual distributions. In contrast with the independent, identically distributed or ARMA settings, these studies illuminate some interesting features of GARCH residuals.

• 응용통계연구
• /
• 제8권2호
• /
• pp.121-132
• /
• 1995

정상조건에서 수명이 상당히 긴 다수의 부품으로 구성된 직렬형 시스템의 수명검사를 현실적으로 수행하기 위해 부분적 단계충격 수명검사의 최적 검사계획에 관하여 고찰하였다. 시스템을 구성하고 있는 부품의 수명이 서로 독립인 지수분포를 따르는 것으로 가정하여 각 부품의 고장률과 가속인자의 최우추정량을 구하였다. 또한 각 부품의 고장률과 가속인자에 관한 최우추정량의 일반화 점근분산의 합과 각 부품의 가속인자에 관한 최우추정량의 점근분산의 합을 구하여 이를 최소가 되게 하는 최적변환시점의 결정방법을 제안하였다.

• Journal of the Korean Statistical Society
• /
• 제28권4호
• /
• pp.415-433
• /
• 1999

This paper considers the design of step-stress accelerated life tests for the Weibull distribution with a nonconstant shape parameter under Type I censoring. It is assumed that scale and shape parameters are log-linear functions of (possibly transformed) stress and that a cumulative exposure model holds for the effect of changing stress. The asymptotic variance of the maximum likelihood estimator of a stated quantile at design stress is used as an optimality criterion. The optimum three step-stress plans are presented for selected values of design parameters and the effects of errors in pre- estimates of the design parameters are investigated.

• Journal of the Korean Statistical Society
• /
• 제26권1호
• /
• pp.101-116
• /
• 1997

This is primarily an expository paper that presents several nonparametric procedures for testing exponentiality against certain monotonicity properties of the mean residual life function, tests against the trend change in such function attract a great deal of attention of late in reliability analysis. In this note, we present some of the known testing procedures regarding the behavior of mean residual life function. These tests are also compared in terms of asymptotic relative efficiency and empirical power against a few alternatives. The tests based on incomplete data are also briefly discussed.

• Journal of the Korean Statistical Society
• /
• 제35권2호
• /
• pp.115-142
• /
• 2006

This paper studies the problem of option hedging when the underlying asset price process is a compound Poisson process. By adopting an asymptotic approach to let the security price converge to a continuous process, we find a closed-form hedging strategy that improves the classical Black-Scholes hedging strategy in a quadratic sense. We first show that the scaled Black-scholes hedging error has a limit in law, and that limit is decomposed into a part that can be traded away and a part that is purely unreplicable. The Black-Scholes hedging strategy is then modified by adding the replicable part of its hedging error and by adding the mean-variance hedging strategy to the nonreplicable part. Some results of simulation experiment s are also provided.

• Journal of the Korean Statistical Society
• /
• 제26권1호
• /
• pp.31-46
• /
• 1997

Let $X_{t}$ = .beta. $X_{{t-1}}$ + .epsilon.$_{t}$ be an autoregressive process where $\mid$.beta.$\mid$ < 1 and {.epsilon.$_{t}$} is independent and identically distriubted with regularly varying tail probabilities. This process is called the asymptotically stationary first-order autoregressive process (AR(1)) with infinite variance. In this paper, we obtain a host of weak convergences of some point processes based on bootstrapping of { $X_{t}$}. These kinds of results can be generalized under the infinite variance assumption to ensure the asymptotic validity of the bootstrap method for various functionals of { $X_{t}$} such as partial sums, sample covariance and sample correlation functions, etc.ions, etc.