• The Korean Journal of Applied Statistics
• /
• v.22 no.2
• /
• pp.341-353
• /
• 2009

We consider sample-size determination problem motivated by comparative clinical trials where patient outcomes are characterized by a bivariate outcome of efficacy and safety. Thall and Cheng (1999) presented a sample size methodology for the case of bivariate binary outcomes. We propose a bivariate Wilcoxon-Mann-Whitney(WMW) statistics for sample-size determination for binary outcomes, and this nonparametric method can be equally used to determine sample sizes of ordinal outcomes. The two methods of sample size determination rely on the same testing strategy for the target parameters but differs in the test statistics, an asymptotic bivariate normal statistic of the transformed proportions in Thall and Cheng (1999) and nonparametric bivariate WMW statistic in the other method. Sample sizes are calculated for the two experimental oncology trials, described in Thall and Cheng (1999), and for the first trial example the sample sizes of a bivariate WMW statistic are smaller than those of Thall and Cheng (1999), while for the second trial example the reverse is true.

• The Korean Journal of Applied Statistics
• /
• v.16 no.1
• /
• pp.55-69
• /
• 2003

In this article, we propose the weighted LAD (least absolute deviations) estimators for multi-dimensional contingency tables and drive an estimation method to estimate the proposed estimators. To illustrate the robustness of the estimators, simulation results are presented for several models Including log-linear models and models for ordinal variables in multidimensional contingency tables. Examples were also introduced.

• The Korean Journal of Applied Statistics
• /
• v.17 no.3
• /
• pp.459-473
• /
• 2004

It has well known that an exhaustive search algorithm suggested by Breiman et. a1.(1984) has a trend to select the variable having relatively many possible splits as an splitting rule. We propose an algorithm to overcome this variable selection bias problem and then construct unbiased regression trees based on the algorithm. The proposed algorithm runs two steps of selecting a split variable and determining a split rule for binary split based on the split variable. Simulation studies were performed to compare the proposed algorithm with Breiman et a1.(1984)'s CART(Classification and Regression Tree) in terms of degree of variable selection bias, variable selection power, and MSE(Mean Squared Error). Also, we illustrate the proposed algorithm with real data sets.

• The Korean Journal of Applied Statistics
• /
• v.27 no.5
• /
• pp.833-842
• /
• 2014

Non-inferiority trials indicate whether the effect of an experimental treatment is not worse than an active control. Chen et al. (2006) and Kang (2010) proposed a test method for non-inferiority trials using confidence intervals. In this paper, we suggest a new nonparametric method using a confidence interval based on Wilcoxon rank-sum test and Hodges-Lehmann estimator of active control. A Monte-Carlo simulation study compares the type I error and the power of the proposed method with previous methods.

• The Korean Journal of Applied Statistics
• /
• v.27 no.1
• /
• pp.1-11
• /
• 2014

The Weighted Logrank test and its special case, Logrank test are widely used to compare survival distributions; however, these methods are inappropriate when the sample size is small or censoring distributions are not equal since they use test statistics from approximate distributions. A permutation test can be an alternative for small sample cases; however, this should be used only when censoring distributions are equal. To handle cases with small sample size and unequal censoring distributions, the permutation-imputation method was developed to compare two survival distributions. In this paper, approximate method, permutation method and permutation-imputation method were compared using a Logrank test and Prentice-Wilcoxon test for three or more survival distributions comparison.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.423-423
• /
• 2021

국내에서는 '물손실(누수)을 얼마나 잘 관리하고 있는지' 평가하는 물손실관리 수행능 지표를 '유수율(누수율)' 로 판단하고 있다. 하지만, 유수율은 실제 관리자의 의지와 무관한 인자에 영향을 받을 수 있으며, 지역의 여건상 유수율이 높게 나올 수밖에 없는 지역이 있는가 하면 반대로 지역의 여건 상 유수율이 낮게 나올 수밖에 없는 지역이 있다. 따라서 유수율만을 이용하여 물손실관리 수행능력을 평가하는 지표로는 미흡함이 있는 것으로 판단될 뿐만 아니라 IWA에서는 전통적인 백분율을 이용한 지표(손실율, 유수율, 무수율 등)의 문제를 언급하면서 수자원의 효율적인 사용 혹은 상수도사업의 경제성을 나타내는 목적 외에는 사용하지 말아야 한다고 권고하였다. 현재까지의 연구를 보면, 국내 실정에 맞는 다양한 물손실관리 수행능 지표들이 언급되면서 이에 대한 적합성 및 적용성에 대한 많은 연구들이 진행되어 왔다. 기존의 연구들은 지자체 단위로 물손실관리 수행능 지표에 대한 연구를 진행하였다면, 본 연구에서는 소블럭 단위로 수도통계자료를 분석하여 물손실관리를 위한 어떠한 지표를 이용하는 것이 타당한 방법인지 제시하고자 하였다. 비교분석을 하기 위한 지표인 소블럭별 유수율, 실손실율(누수율), 1인당 실손실량, 급수전당 실손실량, 누수평가지표(ILI)를 산정하고, 그 결과에 따라 소블럭별 순위를 결정하였다. 유수율과 급수인구를 고려한 1인당 실손실량과의 비교분석에서 높은 R² (결정계수)값과 소블럭별 순위변동 폭이 적은 것으로 나타났으나, 유수율과 급수전수를 고려한 급수전당 실손실량 및 배수관의 연장과 급수전수, 평균압력을 고려한 누수평가지표와의 비교분석에서는 낮은 R²값과 소블럭별 순위 변동폭이 큰 것으로 나타났다. 앞서 언급한 것처럼, 유수율만을 이용하여 물손실관리 수행능을 설명하기에는 부족할 것으로 판단되며, 유수율 외에도 다양한 지표들을 종합적으로 고려했을 때 가장 바람직한 물손실관리 수행능 평가가 될 것으로 사료된다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.39 no.2
• /
• pp.69-80
• /
• 2002

In this paper, signed-rank based nonparametric detectors are used for direct sequence code division multiple access pseudo-noise code acquisition systems in frequency-selective Rician fading channels. We first derive the locally optimum rank detector, and then propose the locally suboptimum rank (LSR) and k-th order modified signed-rank (MSRk) detectors using approximate score functions. We compare the serial and hybrid parallel double-dwell schemes using the LSR and MSRk detectors with those using the conventional squared-sum (SS) using the cell averaging constant false alarm rate processor and modified sign detectors. From the simulation results, it is shown that the LSR and MSRk detectors perform better than the SS detector using the cell averaging constant false alarm rate processor.

• The Korean Journal of Applied Statistics
• /
• v.31 no.2
• /
• pp.241-251
• /
• 2018

The analysis of gene sets in microarray has advantages in interpreting biological functions and increasing statistical powers. Many statistical methods have been proposed for detecting significant gene sets that show relations between genes and phenotypes, but there is no consensus about which is the best to perform gene sets analysis and permutation based tests are considered as standard tools. When many gene sets are tested simultaneously, a large number of random permutations are needed for multiple testing with a high computational cost. In this paper, several parametric approximations are considered as alternatives of the permutation distribution and the moment based gene set test has shown the best performance for providing p-values of the permutation test closely and quickly on a general framework.

• The Korean Journal of Applied Statistics
• /
• v.29 no.3
• /
• pp.471-485
• /
• 2016

Least squares (LS) regression is a classic method for regression that is optimal under assumptions of regression and usual observations. However, the presence of unusual data in the LS method leads to seriously distorted estimates. Therefore, various robust estimation methods are proposed to circumvent the limitations of traditional LS regression. Among these, there are M-estimators based on maximum likelihood estimation (MLE), L-estimators based on linear combinations of order statistics and R-estimators based on a linear combinations of the ordered residuals. In this paper, robust regression estimators with high breakdown point and/or with high efficiency are compared under several simulated situations. The paper analyses and compares distributions of estimates as well as relative efficiencies calculated from mean squared errors (MSE) in the simulation study. We conclude that MM-estimators or GR-estimators are a good choice for the real data application.

• The Korean Journal of Applied Statistics
• /
• v.28 no.3
• /
• pp.573-581
• /
• 2015

We study problems of sample size determination for one-sample location tests. A simulation study shows that sample size calculations based on approximated distribution do not achieve the nominal level of power. We investigate sample size determinations based on exact distribution and with a power that attains the nominal level.