• 응용통계연구
• /
• 제18권3호
• /
• pp.655-660
• /
• 2005

두 그룹 판별분석에서 자주 사용되는 Rao의 추가정보경정법의 적용에서 변수들의 모든 조합중에 일부분만을 고려하여 변수를 선택하는 방법을 제시하고, 예를 들어 제시된 방법의 효용성을 살펴본다.

• Communications for Statistical Applications and Methods
• /
• 제6권3호
• /
• pp.893-900
• /
• 1999

We use statistical tests which are useful for two-sample problem for detecting edges in gray-level images. An edge is detected by examining changes in gray-level value between adjacent pixel neighborhoods. Some experimental results show that nonparametric detectors such as Mann-Whitney test median test and Kolmogorov-Smirnov test perform effectively in both noisy and noise-free images while parametric T test is sensitive to noise.

• Journal of the Korean Data and Information Science Society
• /
• 제28권4호
• /
• pp.947-955
• /
• 2017

A competing risk analysis has been applied when subjects experience more than one type of end points. Geskus (2011) showed three types of estimators of CIF are equivalent under left truncated and right censored data. We extend his approach to an interval censored competing risk data by using a modified risk set and evaluate their performance under several sample sizes. These estimators show very similar results. We also suggest a test statistic combining Sun's test for interval censored data and Gray's test for right censored data. The test sizes and powers are compared under several cases. As a real data application, the suggested method is applied a data where the feasibility of the vaccine to HIV was assessed in the injecting drug uses.

• 응용통계연구
• /
• 제20권2호
• /
• pp.323-332
• /
• 2007

이 논문에서는 이산코사인변환에 의해 유도된 주파수 공간상에서의 포트맨토검정법을 소개한다. 정상시계열의 경우 이산코사인변환 계수는 점근적으로 독립이고 분산은 자기공분산의 선형결합으로 표시된다. 백색잡음에 대한 이산코사인변환 계수의 공분산 행렬은 모든 대각원소가 시계열의 분산인 대각행렬이다. 시계열의 독립성을 검정하기 위해 계수들을 주파수 영역에 따라 2 또는 3개의 그룹으로 분할하고 그룹간의 분산을 비교하여 자료가 백색잡음인지 아닌지를 검정한다. 또한 계수의 제곱값이 반응변수이고 주파수 대역이 설명변수인 회귀모형에서 기울기를 검정하여 백색잡음 여부를 알아본다. 모의실험 결과를 보면 제안한 검정방법이 대부분의 경우 Ljung-Box 검정보다 높은 검정력을 가지는 것으로 나타났다.

• 품질경영학회지
• /
• 제15권2호
• /
• pp.50-54
• /
• 1987

A distribution-free rank test for parallelism of regression lines against ordered alternatives is considered. The proposed test statistic is based on the Kepner-Robinson's transformation. The null distribution of the proposed statistic is the same as that of the Wilcoxon signed rank statistic. But, the proposed procedure can be applied only to four or fewer regression lines. The results of a small-sample Monte Carlo study show that the proposed test is comparable with the parametric test in heavy tailed distributions.

• Interdisciplinary Bio Central
• /
• 제2권2호
• /
• pp.4.1-4.6
• /
• 2010

The most common type of microarray experiment has a simple design using microarray data obtained from two different groups or conditions. A typical method to identify differentially expressed genes (DEGs) between two conditions is the conventional Student's t-test. The t-test is based on the simple estimation of the population variance for a gene using the sample variance of its expression levels. Although empirical Bayes approach improves on the t-statistic by not giving a high rank to genes only because they have a small sample variance, the basic assumption for this is same as the ordinary t-test which is the equality of variances across experimental groups. The t-test and empirical Bayes approach suffer from low statistical power because of the assumption of normal and unimodal distributions for the microarray data analysis. We propose a method to address these problems that is robust to outliers or skewed data, while maintaining the advantages of the classical t-test or modified t-statistics. The resulting data transformation to fit the normality assumption increases the statistical power for identifying DEGs using these statistics.

• Journal of the Korean Data and Information Science Society
• /
• 제10권2호
• /
• pp.387-395
• /
• 1999

본 연구의 목적은 엔트로피 추정량에 기초한 적합도검정 통계량들을 제시하고 경험적 분포에 기초한 적합도 검정 통계량들과의 검정력을 비교하여 향후 적합도 검정에 사용될 검정력을 선정하는데 그 목적이 있다. 또한 모란 (Moran)의 검정통계량과 엔트로피의 추정을 연구함으로 그 일치성을 알 수 있다.

• 한방재활의학과학회지
• /
• 제20권4호
• /
• pp.105-130
• /
• 2010

Objectives : The purpose of this study was to assess the statistical methods errors used in the journal of Oriental Rehabilitation Medicine(JORM) and to identify the types of errors in statistical analysis. Methods : We reviewed quantitative articles that were published in the JORM from January 2005 through October 2009. Those were not used by statistical analysis such as literature studies, case study, review articles were not included in this analysis. A total of 296 articles was reviewed. We evaluated the adequacy and the validity of the statistical techniques with our checklist established be modified Lee's checklist, and three statistical evaluators assessed together to minimize bias. Results : Of the 222 articles, 213 were used in inferential and descriptive statistics. Of those 80% of articles adopting descriptive and inferential statistics were detected having statistical errors. One articles used 1.7 statistical method unit generally. Most frequently employed statistics were student t-test, one way ANOVA. pearson correlation analysis, Mann-whitney U test, paired t-test, and chi-square test in their order. However, most frequent statistics having errors were similar in order. The most common statistic errors were as follow: 1. absence of normality test, 2. misuse between paired test and unpaired test, 3. wrong choice of repeated measures analysis without consideration of time variables, 4, increase of Type I error by using inappropriate multiple test, 5. inappropriate application of discrete or categorical data instead of continuous data in correlation analysis, 6. poor consideration of basic consumption in chi-square test, 7. confusion between frequency comparison and average comparison, 8. mentioning the statistical technique without using it. Conclusions : We found various mistake or misuses in the applications of statistical methodologies in the articles published in the JORM. Careful consideration of statistical use and review from the specialist of statistics are warranted for improving the quality of JORM.

• Journal of the Korean Data and Information Science Society
• /
• 제20권2호
• /
• pp.261-272
• /
• 2009

신용평가모형에서 두 집단의 판별력 검정방법 중의 하나로 두 분포함수의 동일성 검정을 위한 비모수적인 Kolmogorov-Smirnov (K-S) 검정방법이 대표적으로 적용되고 있다. 본 연구에서는 신용평가모형에서 두 분포함수의 동일성 검정을 위하여 K-S 검정 방법 외에 Cramer-Von Mises, Anderson-Darling, Watson 검정방법들을 소개하고 Joseph (2005)의 기준에 대응하는 판단기준을 제안한다. 또한 신용평가 자료와 유사한 상황 하에서의 모의실험을 통해서 불량률, 표본크기 그리고 제II종 오류율을 고려한 대안적인 판단기준을 제시하고 그 적용방법에 대해서 살펴본다.

• Journal of applied mathematics & informatics
• /
• 제26권3_4호
• /
• pp.679-688
• /
• 2008

In multivariate regression analysis there exist many influence measures on the regression estimates. However it seems to be few of influence diagnostics on test statistics in hypothesis testing. Case-deletion approach is fundamental for investigating influence of observations on estimates or statistics. Tang and Fung (1997) derived single case-deletion of the Wilks' ratio, Lawley-Hotelling trace, Pillai's trace for testing a general linear hypothesis of the regression coefficients in multivariate regression. In this paper we derived more extended form of those measures to deal with joint influence among observations. A numerical example is given to illustrate the effect of joint influence on the test statistics.