• 품질경영학회지
• /
• 제31권4호
• /
• pp.164-175
• /
• 2003

This paper investigates bootstrap confidence intervals of the process capability index(PCI) based on the expected loss derived from the empirical distribution function(EDF). The PCI based on the expected loss is too complex to derive its confidence interval analytically, so the bootstrap method is a good alternative. We propose three types of the bootstrap confidence interval; the standard bootstrap(SB), the percentile bootstrap(PB), and the acceleration biased­corrected percentile bootstrap(ABC). We also perform a comprehensive simulation study under various process distributions, in order to compare the accuracy of the coverage probability of the bootstrap confidence intervals. In most cases, the coverage probabilities of the bootstrap confidence intervals from the EDF PCI turned out to be more accurate than those from the PCI based on the normal distribution. It is expected that the bootstrap confidence intervals from the EDF PCI can be utilized in real processes where the true distribution family may not be known.

• Communications for Statistical Applications and Methods
• /
• 제24권1호
• /
• pp.1-13
• /
• 2017

This manuscript summarized advances in bootstrap methods for long-range dependent time series data. The stationary linear long-memory process is briefly described, which is a target process for bootstrap methodologies on time-domain and frequency-domain in this review. We illustrate time-domain bootstrap under long-range dependence, moving or non-overlapping block bootstraps, and the autoregressive-sieve bootstrap. In particular, block bootstrap methodologies need an adjustment factor for the distribution estimation of the sample mean in contrast to applications to weak dependent time processes. However, the autoregressive-sieve bootstrap does not need any other modification for application to long-memory. The frequency domain bootstrap for Whittle estimation is provided using parametric spectral density estimates because there is no current nonparametric spectral density estimation method using a kernel function for the linear long-range dependent time process.

• Communications for Statistical Applications and Methods
• /
• 제13권3호
• /
• pp.745-754
• /
• 2006

It has been shown that LAD estimates are more efficient than LS estimates when the error distribution is double exponential in AR(1) model. In order to explore the performance of LAD estimates one can use bootstrap approaches. In this paper we consider the efficiencies of bootstrap methods when we apply LAD estimates with highly variable data. Monte Carlo simulation results are given for comparing generalized bootstrap, stationary bootstrap and threshold bootstrap methods.

• 산업경영시스템학회지
• /
• 제30권2호
• /
• pp.37-43
• /
• 2007

ANOVA is widely used for measurement system analysis. It assumes that the measurement error is normally distributed, which may not be seen in certain industrial cases. In this study, the exact and bootstrap confidence intervals for precision-to-tolerance ratio (PTR) are obtained for the cases where the measurement errors are normally and non-normally distributed and the reproducibility variation can be ignored. Lognormal and gamma distributions are considered for non-normal measurement errors. It is assumed that the quality characteristics have the same distributions of the measurement errors. Three different bootstrap methods of SB (Standard Bootstrap), PB (Percentile Bootstrap), and BCPB (Biased-Corrected Percentile Bootstrap) are used to obtain bootstrap confidence intervals for PTR. Based on a coverage proportion of PTR, a comparative study of exact and bootstrap methods is performed. Simulation results show that, for non-normal measurement error cases, the bootstrap methods of SB and BCPB are superior to the exact one.

• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.551-562
• /
• 1995

We consider the iterated bootstrap method in regression model with heterogeneous error variances. The iterated wild bootstrap confidence intervla of the mean response is considered. It is shown that the iterated wild bootstrap confidence interval has coverage error of order $n^{-1}$ wheresa percentile method interval has an error of order $n^{-1/2}$. The simulation results reveal that the iterated bootstrap method calibrates the coverage error of percentile method interval successfully even for the small sample size.

• 한국시뮬레이션학회:학술대회논문집
• /
• 한국시뮬레이션학회 2001년도 춘계 학술대회 논문집
• /
• pp.169-169
• /
• 2001

With the aid of fast computing power, resampling techniques are being introduced for simulation output analysis (SOA). Autocorrelation among the output from discrete-event simulation prohibit the direct application of resampling schemes (Threshold bootstrap, Binary bootstrap, Stationary bootstrap, etc) extend its usage to time-series data such as simulation output. We present a new method for inference from a regenerative process, regenerative bootstrap, that equals or exceeds the performance of classical regenerative method and approximation regeneration techniques. Regenerative bootstrap saves computation time and overcomes the problem of scarce regeneration cycles. Computational results are provided using M/M/1 model.

• Communications for Statistical Applications and Methods
• /
• 제20권1호
• /
• pp.85-96
• /
• 2013

A new bootstrap method combined with the stationary bootstrap of Politis and Romano (1994) and the classical residual-based bootstrap is applied to stationary autoregressive (AR) time series models. A stationary bootstrap procedure is implemented for the ordinary least squares estimator (OLSE), along with classical bootstrap residuals for estimated errors, and its large sample validity is proved. A finite sample study numerically compares the proposed bootstrap estimator with the estimator based on the classical residual-based bootstrapping. The study shows that the proposed bootstrapping is more effective in estimating the AR coefficients than the residual-based bootstrapping.

• Communications for Statistical Applications and Methods
• /
• 제20권1호
• /
• pp.41-52
• /
• 2013

The stationary bootstrap of Politis and Romano (1994) is adopted to develop prediction intervals of returns and volatilities in a generalized autoregressive heteroskedastic (GARCH)(p, q) model. The stationary bootstrap method is applied to generate bootstrap observations of squared returns and residuals, through an ARMA representation of the GARCH model. The stationary bootstrap estimators of unknown parameters are defined and used to calculate the stationary bootstrap samples of volatilities. Estimates of future values of returns and volatilities in the GARCH process and the bootstrap prediction intervals are constructed based on the stationary bootstrap; in addition, asymptotic validities are also shown.

• 한국시뮬레이션학회논문지
• /
• 제19권3호
• /
• pp.119-125
• /
• 2010

시뮬레이션 출력 분석 방법인 이동 블록 부트스트랩이나 정상 부트스트랩, 그리고 임계값 부트스트랩은 자기상관성이 존재하는 데이터에 적용 가능한 표본 재추출 방법론들이다. 이러한 부트스트랩 방법들은 데이터의 정상성을 가정하여 적용해 왔다. 그러나 실제 자료 또는 시뮬레이션 출력에 계절성이나 추세를 동반하여 그 정상성을 보장할 수 없는 경우에는 부트스트랩을 시뮬레이션 출력 분석에 적용하지 못하였다. 시뮬레이션 출력 분석 기법 중 자기상관성을 가장 잘 묘사하는 방법은 임계값 부트스트랩 방법이다. 임계값 부트스트랩은 자료의 임계값을 기준으로 주기를 형성하여 재추출하는 방법으로써 계절성이 존재하는 데이터에 부트스트랩을 적용한다면 임계값 부트스트랩과 유사한 정확도를 얻을 수 있다. 본 논문에서는 계절성이 존재하는 시계열 자료에 대한 부트스트랩 적용 가능성을 제시 및 검증해보고자 한다.

• 한국수자원학회논문집
• /
• 제35권3호
• /
• pp.275-284
• /
• 2002

본 연구에서는 Monte-Carlo 모형, AR(1)모형, PAR(1) 모형과 같은 추계학적 모형의 잔차값을 무작위적 복원추출하여 연 및 월 하천 유출량자료를 모의발생하였다. Bootstrap이라고 불리우는 이 복원추출방법은 자료의 모집단의 가정이 필요없다는 장점이 있으며 자료로부터 직접 통계적 분포형을 추정하는 방법으로써 자료의 순위변동법을 이용한다. 본 연구에서는 이 방법을 용담지점에 적용하였으며 Bootstrap 방법으로 모의발생된 하천 유출량자료의 거동을 검토하기 하기 위해 관측 유출량과 모의 발생된 유출량의 통계치를 산정하여 비교하였다. 그 결과 기존의 방범과 Bootstrap 방법 모두 평균, 표준편차, 자기상관성은 잘 재현하였으나 왜곡도 계수의 경우 Bootstrap 방법이 더 뛰어남을 확인할 수 있었다.