• The Korean Journal of Applied Statistics
• /
• v.29 no.4
• /
• pp.699-706
• /
• 2016

Outlier detection methods without performing a test often do not succeed in detecting multiple outliers because they are structurally vulnerable to a masking effect or a swamping effect. This paper considers testing procedures supplemented to a clustering-based method of identifying the group with a minority of the observations as outliers. One of general steps is performing a variety of t-test on individual outlier-candidates. This paper proposes a sequential procedure for searching for outliers by changing cutoff values on a cluster tree and performing a test on a set of outlier-candidates. The proposed method is illustrated and compared to existing methods by an example and Monte Carlo studies.

• Journal of the Korean Data Analysis Society
• /
• v.20 no.6
• /
• pp.2805-2812
• /
• 2018

The methylation score, expressed as a percentage of the methylation status data derived from the iterative sequencing process, has a value between 0 and 1. It is contrary to the assumption of normal distribution that simply applying the t-test to examine the difference in population-specific methylation scores in these data. In addition, since the result may vary depending on the number of repetitions of sequencing in the process of methylation score generation, a method that can analyze such errors is also necessary. In this paper, we introduce the symbolic data analysis and the interval K-S test method which convert observation data into interval data including uncertainty rather than one numerical data. In addition, it is possible to analyze the characteristics of methylation score by using Beta distribution without using normal distribution in the process of converting into interval data. For the data analysis, the nature of the proposed method was examined using sequencing data of actual patients and normal persons. While the t-test is only possible for the location test, it is found that the interval type K-S statistic can be used to test not only the location parameter but also the heterogeneity of the distribution function.

• Journal of the Korea Computer Industry Society
• /
• v.9 no.5
• /
• pp.223-228
• /
• 2008

A case study to show MS Excel's convenient and powerful functions was conducted to test hypotheses with subway data. Quantitative variables were described using descriptive menu, and qualitative variables were described using histogram menu of a MS Excel software. Relationships were tested using regression menu, differences were tested using t-test menu, and factors were tested using variance-layout menu of a Excel software. Data input, management, and statistical analysis were done successfully with only a MS Excel software.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1243-1251
• /
• 2010

For clinical trials, it is common to compare the placebo and new drug. The method of calculating a sample size for two independent populations are the t-test that is used for parametric methods, and the Wilcoxon rank-sum test that is used in the non-parametric methods. In this paper, we propose a method that is using Kim's (1994) statistic power based on the linear placement statistic, which was proposed by Orban and Wolfe (1982). We also compare the sample size for the proposed method with that for using Wang et al. (2003)'s sample size formula which is based on Wilcoxon rank-sum test, and with that of t-test for parametric methods.

• The Korean Journal of Applied Statistics
• /
• v.20 no.1
• /
• pp.147-157
• /
• 2007

Edge detection has been widely used in computer vision and image processing. We describe a new edge detector based on the robust rank-order test which is a useful alternative to Wilcoxon test. Our method is based on detecting pixel intensity changes between two neighborhoods with a $r{\times}r$ window using an edge-height model to perform effectively on noisy images. Some experiments of our robust rank-order detector with several existing edge detectors are carried out on both synthetic images and real images with and without noise.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.23 no.3
• /
• pp.251-260
• /
• 2005

The Development of CCD has contributed to great advancement in mapping technology with giving benefits to research community of photogrammetry. The purpose of this paper is to find the best selection of interpolation method for creating a terrain model form close range digital photogrammetry. T-test as a kind of statistical analysis was conducted to analyze the similarity of hydraulic model with close range digital photogrammetry and trigonometric leveling. Also, many interpolation methods such as inverse distance, kriging, nearest neighbor and TIN about the hydraulic model interpolation were conducted to compare the results for computer to display actual terrain an optimum interpolation of the digital elevation model form close range digital photogrammetry. The results revealed that kriging and TIN interpolation were efficient methods to judge the hazard interpolation law by analyzing geometric similarity of hydraulic model against hydraulic model application.

• Proceedings of the Korean Reliability Society Conference
• /
• 2000.04a
• /
• pp.185-191
• /
• 2000

A survival variable is a nonnegative random variable X with distribution function F and a survival function (equation omitted)=1－F. This variable is said to be New Better than Used of specified age $t_{0}$ if (equation omitted) for all $\chi$$\geq$0 and a fixed to. We propose the test for $H_{0}$ : (equation omitted) for all $\chi$$\geq$0 against $H_1$：(equation omitted) for all $\chi$$\geq$0 when the specified age $t_{0}$ is unknown but can be estimated from the data when $t_{0}$=${\mu}$, the mean of F, and also when $t_{0}$=$\xi_p$, the pth percentile of F. This test statistic, which is based on a linear function of the order statistics from the sample, is readily applied in the case of small sample. Also, this test statistic is more simple than the test statistic of Ahmad's test statistic (1998). Finally, the performance of this test is presented.

• Electronics and Telecommunications Trends
• /
• v.8 no.4
• /
• pp.157-162
• /
• 1993

본 연구는 아날로그 신호를 사용하는 공중교환 전화망과 접속되는 그룹 3(G3) 팩시밀리의 화상 품질을 측정하는 방법을 제시하였다. CCITT(현 ITU-TS) 표준시험 도표 No.2를 이용하여 전송된 화상에 대한 평가는 설문조사를 통해 평가되었고, 그것들은 MOS(Mean Opinion Score) 방법에 의해 계량화되었다. 설문지의 결과에 대한 상관 분석을 통해 문항을 하나의 종합 평가 문항으로 줄일 수 있음을 살펴보았다. 그리고 그 점수들의 평균들에 대한 차이를 분석함으로써 팩시밀리 화상 품질에 영향을 미치는 요인들의 유의성을 검정하였다. 유의성을 검정하는 방법들로 t 검정법과 Vander Waerden Scores 방법을 제시하였다. 그리고 검정 결과 점수 평균이 유의하지 않은 그룹들을 하나의 그룹으로 하여 그 그룹에 있어서 점수 히스토그램을 구하였다. 이 히스토그램을 하나의 정규 분포 곡선으로 근사시켜 팩시밀리 화상 품질 평가치를 살펴보았다.

• Journal of Korean Society for Quality Management
• /
• v.29 no.2
• /
• pp.37-45
• /
• 2001

A survival variable is a non-negative random variable X with distribution function F(t) satisfying F(0) : 0 and a survival function F(t): 1-F(t). This variable is said to be New Better than Used of specified age t$_{0}$ if F(x+ t$_{0}$)$\leq$F(x).F(t$_{0}$) for all x$\geq$0 and a fixed t$_{0}$. We propose the test for H$_{0}$ : F(x+t$_{0}$)=F(x).F(t$_{0}$) for all x$\geq$0 against H$_1$: F(x+t$_{0}$) $\leq$ F(x).F(t$_{0}$) for all x$\geq$0 when the specified age to is unknown but can be estimated from the data when t$_{0}$=ζ$_{p}$, the pth percentile of F. This test statistic, which is based on the normalized spacings between the ordered observations, is readily applied in the case of small sample. Also, our test is more simple than Ahmad's test (1998). Finally, the performance of our test is presented.our test is presented.

• The Korean Journal of Applied Statistics
• /
• v.22 no.6
• /
• pp.1355-1364
• /
• 2009

This work propose a robust test for location parameters in multivariate data based on MVE and MCD with the affine equivariance and the high-breakdown properties. We consider the hypothesis testing satisfying high efficiency and high test power simultaneously to bring in the one-step reweighting procedure upon high-breakdown estimators, which generally suffer from the low efficiency and, as a result, usually used only in the exploratory analysis. Monte Carlo study shows that the suggested method retains nominal significance levels and higher testing power without regard to various population distributions than a Hotelling's $T^2$ test. In an example, a data set containing known outliers does not make an influence toward our proposal, while it renders a Hotelling's $T^2$ useless.