• Communications for Statistical Applications and Methods
• /
• 제8권1호
• /
• pp.127-135
• /
• 2001

In this paper, we consider a simple method for testing the assumption of independent censoring on the basis of a Cox proportional hazards regression model with a time-dependent covariate. This method involves a two-stage sampling in which a random subset of censored observations is selected and followed-up until their true survival times are observed. Lee and Wolfe(1998) proposed an adjusted estimate of the survivor function for the dependent censoring under a proportional hazards alternative. This paper extends their result to obtain a bootstrap confidence interval for the adjusted survivor function under the dependent censoring. The proposed procedure is illustrated with an example of a clinical trial for lung cancer analysed in Lee and Wolfe(1998).

• Communications for Statistical Applications and Methods
• /
• 제25권1호
• /
• pp.1-13
• /
• 2018

In many studies, a researcher attempts to describe a population where units are measured for multiple outcomes, or responses. In this paper, we present an efficient procedure based on ranked set sampling to estimate and perform hypothesis testing on a multivariate mean. The method is based on ranking on an auxiliary covariate, which is assumed to be correlated with the multivariate response, in order to improve the efficiency of the estimation. We showed that the proposed estimators developed under this sampling scheme are unbiased, have smaller variance in the multivariate sense, and are asymptotically Gaussian. We also demonstrated that the efficiency of multivariate regression estimator can be improved by using Ranked set sampling. A bootstrap routine is developed in the statistical software R to perform inference when the sample size is small. We use a simulation study to investigate the performance of the method under known conditions and apply the method to the biomarker data collected in China Health and Nutrition Survey (CHNS 2009) data.

• 한국수산과학회지
• /
• 제45권5호
• /
• pp.499-506
• /
• 2012

We estimated the population size and fishing mortality of Pacific cod Gadus macrocephalus during the spawning season in waters off Woipo, Geoje Island, Korea, using a mark-recapture method. We marked and released 51 cod>50 cm in total length; six were recaptured by local fishermen during the period from December 15 to 31, 2009. The estimated population size was ca. 180,000 and the fishing mortality of the exploitable cod was 26%. Although we could assume a closed population due to the short survey period, we evaluated the uncertainty in the estimates by applying bootstrap resampling because the sample size was small. The estimated 95% confidence interval was 94,000-568,000 for the population size and 8-49% for fishing mortality. Our study demonstrated that the application of mark-recapture methods and bootstrap resampling can be useful in stock assessment for fisheries management in Korea, but requires a larger sample size, spatially extensive coverage, and sophisticated mark-recapture models based on a refined sampling design for reliable stock assessment and biological reference points in sustainable cod management.

• Food Science and Biotechnology
• /
• 제17권6호
• /
• pp.1352-1356
• /
• 2008

Bootstrap method, a computer-intensive statistical technique to estimate the distribution of a statistic was applied to deal with uncertainty and variability of the experimental data in stochastic prediction modeling of microbial growth on a chill-stored food. Three different bootstrapping methods for the curve-fitting to the microbial count data were compared in determining the parameters of Baranyi and Roberts growth model: nonlinear regression to static version function with resampling residuals onto all the experimental microbial count data; static version regression onto mean counts at sampling times; dynamic version fitting of differential equations onto the bootstrapped mean counts. All the methods outputted almost same mean values of the parameters with difference in their distribution. Parameter search according to the dynamic form of differential equations resulted in the largest distribution of the model parameters but produced the confidence interval of the predicted microbial count close to those of nonlinear regression of static equation.

• 디지털산업정보학회논문지
• /
• 제9권4호
• /
• pp.239-249
• /
• 2013

The role of software process is getting more important to make good quality softwares. One of the measures to improve the software process is Defect Removal Efficiency(DRE). DRE gives a measure of the development team ability to remove defects prior to release. It is calculated as a ratio of defects resolved to total number of defects found. Software Engineering Levels are usually decided by CMMI Model. The model is designed to help organizations improve their software product and service development, acquisition, and maintenance processes. The score of software engineering levels can be calculated by CMMI model. The levels are composed of the three groups(absent, average, and advanced). This study is to find if there is any difference among the three categories in term of the result of software engineering levels on DRE. We propose One way ANOVA to analyze influence of software engineering levels on DRE. Bootstrap method is also used to estimate the sampling distribution of the original sample because the data are not sampled randomly. The method is a statistical method for estimating the sampling distribution of an estimator by sampling with replacement from the original sample. The data were collected in 106 software development projects by the survey. The result of this study tells that there is some difference of DRE among the groups. The higher the software engineering level of a specific company becomes, the better its DRE gets, which means that the companies trying to improve software process can increase their good management performance.

• Journal of the Korean Data and Information Science Society
• /
• 제14권2호
• /
• pp.247-255
• /
• 2003

Hypothesis testing for the equality of two distributions were considered. Nonparametric kernel density estimates were used for testing equality of distributions. Cross-validatory choice of bandwidth was used in the kernel density estimation. Sampling distribution of considered test statistic were developed by resampling method, called the bootstrap. Small sample Monte Carlo simulation were conducted. Empirical power of considered tests were compared for variety distributions.

• 응용통계연구
• /
• 제13권2호
• /
• pp.415-425
• /
• 2000

붓스트랩 표본추출과 pairwise coupling의 알고리즘을 결합한 새로운 분류 알고리즘을 제안하고, 이를 선형판별분석과 2차 판별분석에 적용하였다. 그리고 새로운 분류 알고리즘의 정확도를 비교하기위해 널리 사용되는 waveform 자료 등을 분석한 후, 그 결과를 기존 분류 방법과 비교하였다.

• Journal of the Korean Statistical Society
• /
• 제35권3호
• /
• pp.331-341
• /
• 2006

An approximate distribution of the plug-in estimator of the squared coefficient of variation ($CV^2$) is derived by using Edgeworth expansions under general population models. Also bias of the estimator is investigated for several important distributions. Under the normal distribution, we proposed the new estimator for $CV^2$ based on median of the sampling distribution of plug-in estimator.

• 응용통계연구
• /
• 제27권4호
• /
• pp.645-653
• /
• 2014

p-값은 관측 표본과 관측 결과보다 심하게 대안가설의 방향으로 영가설을 이탈하는 표본들이 영가설 하에서 갖는 확률이다. p-값이 일정 ${\alpha}$(= 0:05)보다 작게 나타나면 연구자는 대안가설이 지지된 것으로 본다. 그런 경우라고 하더라도 그의 가설이 향후 연구에서 번복될 수 있는데 그 이유는 p-값이 표본에 따라 변동하는 통계량이기 때문이다. Boos와 Stefanski (2011)는 붓스트랩 방법으로 p-값의 예측분포를 구할 수 있음을 보였다. 그들은 그 분포의 상위 10-20% 분위수가 ${\alpha}$보다 작은가를 확인할 필요가 있음을 강조한다. 만약 그렇지 않은 경우에는 "지지"된 가설의 재현성이 문제될 수 있기 때문이다. 가설검정에서 일정 수준의 재현율을 확보하기 위해서는 표본의 증대가 요구된다. 이 연구는 k배 확대 붓스트랩 표본추출(boosted bootstrap sampling)로써 필요한 표본크기를 계산할 수 있음을 두 표본의 비교와 다중선형회귀의 수치 예에서 보인다. k 값을 정하기 위해서는 몇 차례 시행착오를 해야 하지만 계산적 부담은 크지 않다. 95% 신뢰구간은 독립적인 표본들로부터 같은 방식으로 산출되는 구간이 미지의 모수를 포함할 확률이 95%가 되도록 설정된다. 이 연구는 한 관측표본으로부터 얻어진 95% 신뢰구간 내 개별 점이 미래 연구의 신뢰구간에도 포함될 것인지 그 재현성을 붓스트랩 재표본들에서 평가한다. 이 연구는 개별 점에서 산출한 신뢰구간 재현율을 그래프로 보인다.

• 응용통계연구
• /
• 제4권1호
• /
• pp.1-11
• /
• 1991

다변량분산분석이나 판별분석 등에 있어서 검정의 대상이 되는 공분산행렬의 동일성에 대한 붓스트랩방법의 활용을 살펴보았다. 두 모집단의 공분산행렬을 $\Sigma_1, \Sigma_2^$라 하면, 가설 H : $\Sigma_1 = \Sigma_2$은 불변성의 관점에서 $\Sigma = \Sigma_1 \Sigma_2^{-1}$의 고유값들이 모두 1 이라는 것과 동등하다. 본 연구에서는 (1) $\Sigma = \Sigma_1 \Sigma_2^{-1}$의 표본고유값들에 대한 편의를 붓스트랩에 의해 정정하였으며, (2) 이들의 표본분포를 붓스트랩분포로 추정하여 검정에 활용하였으며, (3) 합동붓스트랩에 의해 바플렛의 수정우도비 검정통계량의 분포를 근사하였다.