• Communications for Statistical Applications and Methods
• /
• 제13권1호
• /
• pp.39-47
• /
• 2006

The Bayesian approach to multiple testing procedure for one sample testing problem proposed by Scott and Berger (2003) is extended to two-sample comparison problem in microarray experiments. The prior distribution of each gene's mean for one sample is given conditionally on the corresponding gene's mean for the other sample. Posterior distributions of interesting parameters are derived and estimated based on an importance sampling method. A simulated example is given for illustration.

• 대한간호학회지
• /
• 제30권6호
• /
• pp.1503-1520
• /
• 2000

The purpose of the present study is 1) to explain major effect models (main, moderating, and mediating) of social support and statistical methods for testing the effect models and 2) to analyze and evaluate the consistency in the use of the effect models and its statistical methods in Korean nursing studies. A total of 57 studies were selected from Journal of Korean Academy of Nursing, Journal of Korean Academic Society of Adult Nursing, Journal of Korean Women's Health Nursing Academic Society, Journal of Fundamentals of Nursing, Journal of Korean Community Nursing, Journal of Korean Psychiatric and Mental Health Nursing Academic Society, and Journal of Korean Pediatric Nursing Academic Society published in the year of 1990-1999. In results, most studies on social support performed in Korea Nursing Society were about a main effect model. There are few studies on moderating or mediating model of social support. Thus, it was difficult to find research findings how, why, under what conditions social support impacted on health outcomes. Most studies on the moderating or mediating effect model of social support used statistical methods for testing main effect model rather than for testing moderating or mediating effect model. That is, there are inconsistency between effect models of social support and its statistical methods in Korean nursing researches. Therefore, it is recommended to perform studies on moderating or mediating effect model and use appropriate statistical methods.

• 응용통계연구
• /
• 제2권1호
• /
• pp.35-47
• /
• 1989

일반화된 ABO-식 혈액형 구조에서 나타나는 m개 유전자 빈도에 대한 추정 방법중 최우추정법에 대하여 논하였으며, 이 추정치를 기초로 유전자 빈도의 차이에 대한 동질성 검정문제에 있어서의 유전자 갯수 m이 3 이상인 경우에도 성립하게 되는 일반화를 시도함으로써 m개 유전자 빈도에 대한 검정도 가능하게끔 하였다. 한편 응용 예에서는 최우추정치와 그 외의 다른 방법들-즉 Bernstein 방법, 조정된(Adjusted) Bernstein 방법 그리고 수정된 (Modified) Bernstein 방법등-에 의한 추정치들을 비교 분석하였으며, 직교분할을 기초로 하여, 동질성 문제에 대한 통계적 검정도 실시되었다.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제14권2호
• /
• pp.93-96
• /
• 2014

RFID (Radio Frequency IDentification) system is recognition technology which can maintain various object's information. Reliability of RFID tags is the most important factor in RFID system. In this paper, we proposed ALT (Accelerated Life Test) method for UHF RFID tags. Temperature and humidity were adopted as stress factors and the accelerated life tests were conducted in three different conditions. We performed failure analysis for identifying failure mechanism and statistical analysis of test data. In the statistical analysis, we employed Inverse Power law for relationship between tag's life and stress. Through the statistical analysis, we proposed acceleration factor for several levels of temperature-humidity. The reliability qualification test plans were also designed for the tag's target reliability.

• 정보보호학회논문지
• /
• 제11권6호
• /
• pp.67-76
• /
• 2001

본 논문에서는 미국 NIST(National Institute of Standards & Technology)의 AES(Advanced Encry-ption Standard) 선정기준 중 안전성 평가인 난수검정에 대하여 고찰하고자 한다. 암호 알고리즘의 안전성 평가는 입출력문과 키의 크기, 평문과 암호문 및 키와 암호문의 상관성, 평문과 키의 변화에 따르는 암호문의 변화 그리고 구조적 특이성 등이 고려대상이 된다. 주어진 암호 알고리즘에 대한 안전성 필요충분조건 만족여부를 평가하는 것은 어려우며 객관적인 평가를 위해서는 정량적인 평가결과가 도출되어야 한다. 본 논문에서는 NIST에서 실시한 AES 안전성 평가항목들과 기준에 대하여 고찰하며, 국내 암호 알고리즘 표준인 SEED등 여러 암호 알고리즘과 난수발생기를 AES 평가기준에 맞추어 새롭게 분석해 보고자 한다.

• Journal of the Korean Statistical Society
• /
• 제13권2호
• /
• pp.73-80
• /
• 1984

In this paper, likelihood-ratio test has been derived for testing multisample sphericity in complex multivariate Gaussian populations. The $h^{th}$ moment of the test statistic is given and its exact distribution has been derived using inverse Mellin transform. Asymptotic distribution of the statistic is also given.

• Communications for Statistical Applications and Methods
• /
• 제7권2호
• /
• pp.403-414
• /
• 2000

A diagnostic tool for testing homogeneity for random effects is proposed in unbalanced linear mixed model based on score statistic. The finite sample behavior of the test statistic is examined using Monte Carlo experiments examine the chi-square approximation of the test statistic under the null hypothesis.

• Journal of the Korean Statistical Society
• /
• 제8권2호
• /
• pp.99-105
• /
• 1979

In Thoman and Bain and Schafer and Sheffield, procedures for testing the equality of the scale parameters of two Weibull populations with a common shape parameter and procedures for selecting the Weibull population with the largest scale parameter are given. We give, in this paper, a modified procedure for the above testing and selection problems, which is more powerful than those previoulsy given.

• Journal of the Korean Statistical Society
• /
• 제8권1호
• /
• pp.15-21
• /
• 1979

A zone of sensitivity is developed to investigate the effect of nonnormality on the Bayes decision function for testing mean of a normal population when either parent or prior belongs to Edgeworthian family of moderately nonnormal probability density functions.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2012년 추계학술대회
• /
• pp.461-471
• /
• 2012

Most useful statistical techniques in six sigma DMAIC are hypothesis testing and interval estimation. So this paper reviews and derives sample size formula by considering significance level, power of detectability and effect difference. The quality practioners can effectively interpret the practical and statistical significance with the rational sample sizing.