• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2003.05a
• /
• pp.1179-1185
• /
• 2003

시뮬레이션 상의 입력모델에 대한 기존의 연구는 과거의 자료를 바탕으로 선형의 모수적인 (parametric) 모형을 개발하는데 초점을 두고 있다. 그러나 이 경우에는 입력이 매우 복잡한 형태를 가지면 모수적인 모형을 잦는 것이 불가능해지므로 비모수적인(non-parametric) 접근방법이 절실한 실정이다 예로 인터넷 트래픽 모델의 시뮬레이션 수행시 입력으로 제공되는 단위 시간당 요구되는 웹 페이지의 수 같은 경우 데이터들 간데 종속관계가 매우 심하고 복잡하여 모수적 모형을 세우는데 어려움이 있다. 이러한 시스템들을 시뮬레이션 방법으로 분석 하고자 할 때, 기존의 trace-driven 시뮬레이션 방법이나 모수적 모형을 찾아 다수의 사실적인 시뮬레이션 입력 자료를 확보하는 것은 현실적으로 어려움이 있다. 따라서. 비모수적인 방법으로 다수의 사실적인 시뮬레이션 입력 자료를 생성하는 것이 필요하다. 이러한 비모수적인 방법에 대한 평가기준 설정은 시뮬레이션 상의 입력 모델에 대한 타당성을 제시한다는 점에서 또한 매우 중요하다. 본 논문에서는 붓트스트 랩의 방법중의 하나인 임계값 붓트스트랩을 이용하여 시뮬레이션 입력 자료 생성 방법을 개발하였고 Turing test를 통해 붓스트랩으로 생성산 입력 시나리오를 검증하였다.

• Journal of Veterinary Clinics
• /
• v.24 no.4
• /
• pp.509-513
• /
• 2007

Parametric and nonparametric coupled with bootstrap simulation technique were used to reevaluate previously defined reference intervals of serum chemistry parameters. A population-based study was performed in 100 clinically healthy dogs that were retrieved from the medical records of Kangwon National University Animal Hospital during 2005-2006. Data were from 52 males and 48 females(1 to 8 years old, 2.2-5.8 kg of body weight). Chemistry parameters examined were blood urea nitrogen(BUN)(mg/dl), cholesterol(mg/dl), calcium(mg/dl), aspartate aminotransferase(AST)(U/L), alanine aminotransferase(ALT)(U/L), alkaline phosphatase(ALP)(U/L), and total protein(g/dl), and were measured by Ektachem DT 60 analyzer(Johnson & Johnson). All but calcium were highly skewed distributions. Outliers were commonly identified particularly in enzyme parameters, ranging 5-9% of the samples and the remaining were only 1-2%. Regardless of distribution type of each analyte, nonparametric methods showed better estimates for use in clinical chemistry compare to parametric methods. The mean and reference intervals estimated by nonparametric bootstrap methods of BUN, cholesterol, calcium, AST, ALT, ALP, and total protein were 14.7(7.0-24.2), 227.3(120.7-480.8), 10.9(8.1-12.5), 25.4(11.8-66.6), 25.5(11.7-68.9), 87.7(31.1-240.8), and 6.8(5.6-8.2), respectively. This study indicates that bootstrap methods could be a useful statistical method to establish population-based reference intervals of serum chemistry parameters, as it is often the case that many laboratory values do not confirm to a normal distribution. In addition, the results emphasize on the confidence intervals of the analytical parameters showing distribution-related variations.

• The Korean Journal of Applied Statistics
• /
• v.36 no.6
• /
• pp.561-571
• /
• 2023

Volatility of a time series is defined as the conditional variance on the past information. In particular, for financial time series, volatility is regarded as a time-varying measure of risk for the financial series. To capture the intrinsic asymmetry in the risk of financial series, various asymmetric volatility processes including threshold-ARCH (TARCH, for short) have been proposed in the literature (see, for instance, Choi et al., 2012). This paper proposes a volatility function featuring non-zero origin in which the origin of the volatility is shifted from the zero and therefore the resulting volatility function is certainly asymmetric around zero and achieves the minimum at a non-zero (rather than zero) point. To validate the proposed volatility function, we analyze the Korea stock prices index (KOSPI) time series during the Covid-19 pandemic period for which origin shift to the left of the zero in volatility is shown to be apparent using the minimum AIC as well as via parametric bootstrap verification.

• The Korean Journal of Applied Statistics
• /
• v.15 no.2
• /
• pp.223-232
• /
• 2002

The present paper develops a test of the multivariate normality based on nonlinear transformations and the likelihood function. For checking the normality, we test the shape parameter which indexes the family of transformations. A score test and a parametric bootstrap test are used to evaluate the discrepancy between the data and a multivariate normal distribution. In order to compare the performance of our test with the existing tests, a simulation study was carried out for several situations where nuisance parameters have to be estimated. The results showed that the proposed method is superior to the existing methods.

• Journal of Korean Society for Quality Management
• /
• v.35 no.3
• /
• pp.107-117
• /
• 2007

The fatigue has been considered the most failure mode of metal, ceramic, and composite materials. In this paper, numerical experiments to asses the usefulness of two Dixon's methods(small and large samples) and 14 S-N methods on assumptions of lognormal fatigue limit distribution under RFL(Random Fatigue Limit) model are conducted for staircase(or up-and-down) test and compared by MSE(Mean Squared Error) and bias for estimates of mean log-fatigue limit. Also, guidelines for staircase test plans to choose initial stress level and step size are recommended from numerical experiments including sensitivity analyses. In addition, the parametric bootstrap method to construct a confidence interval for the mean of log-fatigue limit by the percentile method using a transition probability matrix of Markov chain is presented and illustrated with an example.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.25 no.4
• /
• pp.191-199
• /
• 2013

It is estimated and analyzed that the design wave height and the confidence interval (hereafter CI) according to the return period using the fourteen-year wave data obtained at Pusan New Port. The functions used in the extreme value analysis are the Gumbel function, the Weibull function, and the Kernel function. The CI of the estimated wave heights was predicted using one of the Monte-Carlo simulation methods, the Bootstrap method. The analysis results of the estimated CI of the design wave height indicate that over 150 years of data is necessary in order to satisfy an approximately ${\pm}$10% CI. Also, estimating the number of practically possible data to be around 25~50, the allowable error was found to be approximately ${\pm}$16~22% for Type I PDF and ${\pm}$18~24% for Type III PDF. Whereas, the Kernel distribution method, a typical non-parametric method, shows that the CI of the method is below 40% in comparison with the CI of the other methods and the estimated design wave height is 1.2~1.6 m lower than that of the other methods.

• Food Science and Biotechnology
• /
• v.17 no.3
• /
• pp.659-663
• /
• 2008

Different statistical distribution functions were examined to find an adequate distribution function to describe the microbial contamination behavior of a Korean side dish product, seasoned soybean sprouts for different seasons and market groups. The triang distribution was the best for any market groups in winter, while the logistic distribution could describe the microbial contamination in log CFU/g for all the market groups in spring and summer. From parametric bootstrapping based on the fitted distributions, it was found that a normal distribution could describe the distribution of mean microbial count in log CFU/g for all the seasons and market groups. Statistical parameters for each season/market group are presented to estimate the confidence interval.

• The Korean Journal of Applied Statistics
• /
• v.24 no.1
• /
• pp.35-43
• /
• 2011

In this paper, we introduce a unit root test via discrete cosine transform in the AR(1) process. We first investigate the statistical properties of DCT coefficients under the stationary AR(1) process and the random walk process in order to verify the validity of the proposed method. A bootstrapping approach is proposed to induce the distribution of the test statistic under the unit root. We performed simulation studies for comparing the powers of the Dickey-Fuller test and the proposed test.

• Communications for Statistical Applications and Methods
• /
• v.13 no.3
• /
• pp.503-512
• /
• 2006

In this paper, the Inverse Weibull distribution parameters have been estimated using a new estimation technique based on the non-parametric kernel density function that introduced as an alternative and reliable technique for estimation in life testing models. This technique will require bootstrapping from a set of sample observations for constructing the density functions of pivotal quantities and thus the confidence intervals for the distribution parameters. The performances of this technique have been studied comparing to the conditional inference on the basis of the mean lengths and the covering percentage of the confidence intervals, via Monte Carlo simulations. The simulation results indicated the robustness of the proposed method that yield reasonably accurate inferences even with fewer bootstrap replications and it is easy to be used than the conditional approach. Finally, a numerical example is given to illustrate the densities and the inferential methods developed in this paper.

• Communications for Statistical Applications and Methods
• /
• v.28 no.6
• /
• pp.627-641
• /
• 2021

An asymmetric least squares estimation method has been employed to estimate linear models for percentile regression. An asymmetric maximum likelihood estimation (AMLE) has been developed for the estimation of Poisson percentile linear models. In this study, we propose generalized nonlinear percentile regression using the AMLE, and the use of the parametric bootstrap method to obtain confidence intervals for the estimates of parameters of interest and smoothing functions of estimates. We consider three conditional distributions of response variables given covariates such as normal, exponential, and Poisson for three mean functions with one linear and two nonlinear models in the simulation studies. The proposed method provides reasonable estimates and confidence interval estimates of parameters, and comparable Monte Carlo asymptotic performance along with the sample size and quantiles. We illustrate applications of the proposed method using real-life data from chemical and radiation epidemiological studies.