• 대한수학회보
• /
• 제53권2호
• /
• pp.351-363
• /
• 2016

In this paper, we consider the entropy-based goodness of fit test (Vasicek's test) for a composite hypothesis. The test measures the discrepancy between the nonparametric entropy estimate and the parametric entropy estimate obtained from an assumed parametric family of distributions. It is shown that the proposed test is asymptotically normal under regularity conditions, but is affected by parameter estimates. As a remedy, a bootstrap version of Vasicek's test is proposed. Simulation results are provided for illustration.

• 수산경영론집
• /
• 제47권1호
• /
• pp.63-86
• /
• 2016

The purpose of this study is to investigate production efficiency of Oceans and Fisheries Human Resources Development Programs Efficiency using Bootstrap-DEA. The study extracts 33 officials curriculum, 11 fisheries managers curriculum for its analytical. First, the study estimates technical, pure technical, and scale efficiency of each curriculums based on traditional DEA under the assumption of CRS and VRS. 8(official 7, managers 1) curriculums are identified as efficient DMUs under the CCR-model, and 13(official 10, managers 3) under the BCC-model. We provide inputs that allow inefficient curriculum to be efficient DMUs on a production frontier, and a reference set for their bench-marking. Second, rank test, Wilcoxon-Mann-Whitney test to find a statistical significance of heterogeneity existing in efficiences between Bootstrap-DEA tenical vs Bootstrap-DEA pure technical was no significant difference. We have identified that G10, 11, 12 13, 25, 31, 33, 39 curriculums are the most efficiently produced in the technical and pure technical efficiency. Also we managed to measure the inefficiency which exists in efficiently produced curriculums when estimating the bias corrected efficiency scores. In Technical efficiency, Operation and facility was significant at the 10%. In Pure technical efficiency, facility was significant at the 10%.

• Communications for Statistical Applications and Methods
• /
• 제11권3호
• /
• pp.643-655
• /
• 2004

L$_1$-estimator in the linear regression model is widely recognized to have superior robustness in the presence of vertical outliers. While the L$_1$-estimation procedures and algorithms have been developed quite well, less progress has been made with the hypothesis test in the multiple L$_1$-regression. This article suggests computer-intensive resampling approaches, jackknife and bootstrap methods, to estimating the variance of L$_1$-estimator and the scale parameter that are required to compute the test statistics. Monte Carlo simulation studies are performed to measure the power of tests in small samples. The simulation results indicate that bootstrap estimation method is the most powerful one when it is employed to the likelihood ratio test.

• Journal of the Korean Data and Information Science Society
• /
• 제13권1호
• /
• pp.129-137
• /
• 2002

Two-sample problem is frequently discussed problem in statistics. In this paper we consider the hypothese methods for the general two-sample problem and suggest the bootstrap methods. And we show that the modified Kolmogorov-Smirnov test is more efficient than the Kolmogorov-Smirnov test.

• 응용통계연구
• /
• 제29권1호
• /
• pp.61-73
• /
• 2016

본 논문에서는 다변량 시계열 모형 진단을 위해 잔차의 자기상관성 유무를 확인하기 위한 와일드 붓스트랩(wild bootstrap) Ljung-Box(LB) 검정통계량을 연구하였다. 일반적으로 LB 검정은 오차가 서로 독립이며 동일한 분포를 따른다는 IID 가정 하에 유도되는 점근적 카이제곱 분포를 이용한다. 한편 금융시계열 자료는 분산에 조건부 이분산성이 존재하기 때문에 오차의 IID 가정을 만족시키지 못하며 이에 따라 점근적 분포를 이용한 LB 검정은 제1종의 오류를 만족시키지 못하게 된다. 이를 극복하기 위해 와일드 붓스트랩을 이용한 LB 검정법을 제안하고 그 성질을 연구하고자 한다. 벡터자기회귀 모형과 벡터오차수정 모형 등의 다양한 다변량 시계열 모형을 이용하여 모의실험을 실시하는 한편, 코스피 200지수와 지수선물 자료를 이용한 실증분석을 통해 와일드 붓스트랩을 이용한 LB 검정법이 조건부 이분산성의 부정적인 영향을 효과적으로 제거할 수 있음을 입증하였다.

• 품질경영학회지
• /
• 제24권2호
• /
• pp.77-86
• /
• 1996

In this paper, we consider the problem of testing independence in the absolutely continuous bivariate exponential distribution of Block and Basu(1974). We construct a bootstrap procedure for testing zero and non-zero values of the parameter ${\lambda}_3$ which measures the degree of dependence and compare the power of the bootstrap test with likelihood ratio test(LRT) by Gupta et al.(1984) and the test based on maximum likelihood estimator(MLE) $\hat{{\lambda}}_3$ by Hanagal and Kale(1991) for small and moderate sample sizes.

• 한국수자원학회논문집
• /
• 제42권10호
• /
• pp.795-807
• /
• 2009

본 연구에서는 부트스트랩(Bootstrap) 기법을 이용하여 측우기 강우량 관측계열(CWK)과 근대우량계 강우량 관측 계열(MRG)에 대해 동질성 분석을 실시하였다. 서로 다른 두 자료계열에 대한 전통적인 통계적 동질성 검정 방법은 모집단의 분포형을 알고 있어야 검정결과가 유효하였기 때문에 모집단의 분포가 복잡한 기상자료들은 이러한 전통적 방법을 사용하여 동질성을 파악하는 것이 매우 어려웠고 결과로 제시된 통계적 유의성에 대해서도 의심의 여지가 있었다. 이러한 이유로 본 논문에서는 모집단을 가정하지 않아도 되는 비모수적 모의 방법인 부트스트랩 기법을 이용하여 모집단을 직접 추정한 후 경험누가확률분포를 산정하여 두 자료계열간 통계적 동질성 검정을 실시하였다. 분석 결과 CWK와 MRG는 미소한 기후의 경년변화(trend)의 영향을 제외하면 동질성을 가진 자료로 볼 수 있었다.

• Journal of the Korean Statistical Society
• /
• 제27권4호
• /
• pp.495-506
• /
• 1998

The proposed class of generalized logistic models, indexed by an extra parameter, can be used to model or to examine symmetric or asymmetric discrepancies from the logistic model. When there are a finite number of different design points, we are mainly concerned with maximum likelihood estimation of parameters and in deriving their large sample behavior A score test and a bootstrap hypothesis test are also considered to check if the standard logistic model is appropriate to fit the data or if a generalization is needed .

• Journal of the Korean Statistical Society
• /
• 제30권4호
• /
• pp.645-660
• /
• 2001

In this paper we propose a goodness-of-fit test statistic for testing the (null) parametric model versus the (alternative) nonparametric model. Most of existing nonparametric test statistics are based on the residuals which are obtained by regressing the data to a parametric model. Our test is based on the bootstrap estimator of the probability that the smoothing parameter estimator is infinite when fitting residuals to cubic smoothing spline. Power performance of this test is investigated and is compared with many other tests. Illustrative examples based on real data sets are given.

• 산업경영시스템학회지
• /
• 제25권4호
• /
• pp.1-7
• /
• 2002

Statistical process control is used widely as an effective tool to solve the quality problems in practice fields. All the control charts used in statistical process control are parametric methods, suppose that the process distributes normal and observations are independent. But these assumptions, practically, are often violated if the test of normality of the observations is rejected and/or the serial correlation is existed within observed data. Thus, in this study, to screening process, the Combined Shewhart - CUSUM quality control chart is described and evaluated that used bootstrap method. In this scheme the CUSUM chart will quickly detect small shifts form the goal while the addition of Shewhart limits increases the speed of detecting large shifts. Therefor, the CSC control chart is detected both small and large shifts in process, and the simulation results for its performance are exhibited. The bootstrap CSC control chart proposed in this paper is superior to the standard method for both normal and skewed distribution, and brings in terms of ARL to the same result.