• Journal of the Korean Data and Information Science Society
• /
• 제9권2호
• /
• pp.323-336
• /
• 1998

점검표 설계상의 타당성 평가이론에 근거한 점검표 설계 단계별 타당성 평가와 점검표 사용목적에 따른 타당성 평가방법을 제안하여 한국의류학회지를 대상으로 평가이론에 입각하여 점검표의 각 단계별 타당성 평가기준을 설정하였다.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2006년도 International Symposium on GPS/GNSS Vol.1
• /
• pp.35-40
• /
• 2006

For the maritime traffic safety assessment of vessels navigating in harbor or fairway, simulation techniques by using shiphandling simulator system have been traditionally used. When designing the simulation experiments and when analyzing the simulation results, however, there has been a little systematic method. Ship-handling simulations can be regarded as a kind of statistical experiment by using ship-handling simulator system, which means that shiphandling simulation conditions should be designed statistically and that the simulation results should be statistically analyzed as well. For the safe and economic design of harbor and fairway, reasonable decisions based upon the scientific analysis of shiphandling simulation results are indispensable. In this paper, various statistical methods, such as Bayes theorem, statistical hypothesis testing, and probability distributions, are reviewed with a view to application to maritime traffic safety assessment. It is expected that more reasonable decisions on harbor and fairway design can be made from shiphandlers' view point by using statistical methods.

• Communications for Statistical Applications and Methods
• /
• 제29권4호
• /
• pp.453-469
• /
• 2022

The study of immune cellular composition has been of great scientific interest in immunology because of the generation of multiple large-scale data. From the statistical point of view, such immune cellular data should be treated as compositional. In compositional data, each element is positive, and all the elements sum to a constant, which can be set to one in general. Standard statistical methods are not directly applicable for the analysis of compositional data because they do not appropriately handle correlations between the compositional elements. In this paper, we review statistical methods for compositional data analysis and illustrate them in the context of immunology. Specifically, we focus on regression analyses using log-ratio transformations and the alternative approach using Dirichlet regression analysis, discuss their theoretical foundations, and illustrate their applications with immune cellular fraction data generated from colorectal cancer patients.

• Genomics & Informatics
• /
• 제17권4호
• /
• pp.41.1-41.12
• /
• 2019

Survival analysis mainly deals with the time to event, including death, onset of disease, and bankruptcy. The common characteristic of survival analysis is that it contains "censored" data, in which the time to event cannot be completely observed, but instead represents the lower bound of the time to event. Only the occurrence of either time to event or censoring time is observed. Many traditional statistical methods have been effectively used for analyzing survival data with censored observations. However, with the development of high-throughput technologies for producing "omics" data, more advanced statistical methods, such as regularization, should be required to construct the predictive survival model with high-dimensional genomic data. Furthermore, machine learning approaches have been adapted for survival analysis, to fit nonlinear and complex interaction effects between predictors, and achieve more accurate prediction of individual survival probability. Presently, since most clinicians and medical researchers can easily assess statistical programs for analyzing survival data, a review article is helpful for understanding statistical methods used in survival analysis. We review traditional survival methods and regularization methods, with various penalty functions, for the analysis of high-dimensional genomics, and describe machine learning techniques that have been adapted to survival analysis.

• Journal of the Korean Data and Information Science Society
• /
• 제14권2호
• /
• pp.257-264
• /
• 2003

The perception of the importance of statistical methods for processing medical data in Korea's medical research and the practical use of the analysis method are insufficient. From this standpoint, in order to examine the features of the data analysis method used in the medical journals of Korea and America, we have examined the research papers which has been published in the exemplary medical journals of both countries. It showed that there was a large difference in the quantity and quality between Korea and America. Especially in the medical research of Korea, we could notice that the use of statistical methods were comparatively low. Hence the researchers in the medical area are encouraged to use more statistical methods in processing medical data.

• 한방안이비인후피부과학회지
• /
• 제19권3호통권31호
• /
• pp.134-145
• /
• 2006

Objective : This study was carried out to investigate what type of assumption and conditions are needed for the application of various statistical techniques such as descriptive statistics, t-test, analysis of variance, correlation analysis, regression analysis and chi-square test and to evaluate that they are used correctly in the research process. Methods : One more methods of statistical analysis were used in 91 papers among 162 papers selected from the journal of Korean oriental medical Ophthalmology & Otolaryngology & Dermatology from April 2003 to December 2005. So we analysed the type of statistical analysis method in 91 papers(clinical and experimental study) and assessed the their validity of statistical techniques by the check list consisting of 34 items(3 items for validity assessment of descriptive statistics, 6 items for t-test, 7 items for analysis of variance, correlation analysis and regression analysis, respectively, 4 items for chi-square test) Results : 1. The type of 65(40%) cases is experimental trial, the type of 55(34%) cases is case report, the type of 26(16%) cases is clinical trial and the type of 16(10%) cases is review, in 91 papers using statistical techniques among 162 papers selected from the journal of Korean oriental medical Ophthalmology & Otolaryngology & Dermatol-ogy from April 2003 to December 2005. 2. One more methods of statistical analysis were used in the experimental and clinical study. When we classified 125 units using statistical analysis methods in 91 papers according to statistical techniques such as descriptive statistics, t-test, analysis of variance, correlation analysis, regression analysis and chi-square test, the number of independent sample t-test is 33(26%), the number of only descriptive statistics is 28(22%), the number of independent sample t-test is 33(26%), the number of only descriptive statistics is 28(22%), the number of one way ANOVA is 15(12%), the number of non-parametric test 10(8%). 3. After carrying out one way ANOVA, the number of using multiple comparison methods is 15(Scheffe:6(26%), Duncan:4(17%), Dunnett:3(13%), Tukey:2(9%)) out of 23 (total case carrying out one way ANOVA). 8(35%) out of 23 did not enforce multiple comparison methods after carrying out one way ANOVA. 4. From the assessment of validity about 63 cases using statistical techniques(except descriptive statistics), 5(8%) cases are proper, the other 58(92%) are improper, so we recognized a serious misuse of statistical application in our journal. 5. The number of case below 10 sample size in experimental and clinical study(except descriptive statistics) is 31(34%) and frequent. Also the number of case containing no mention of sample size is 41(45%, including culture study). 6. For example of statistical error, there are wrong choice of statistical technique, lack of check on standard assumption(such as standard distribution, equivariance, independence), and so on. Conclusions : We investigated the validity of statistical analysis methods in our journal by check list consisting of 34 items and suggested correct statistical analysis methods. We should practice the spread of education about statistical analysis methods and precis application, enhance objectivity and reliability of our thesis and further correspond with purpose of scientific study.

• Tuberculosis and Respiratory Diseases
• /
• 제57권2호
• /
• pp.168-179
• /
• 2004

배 경 : 의학연구에서는 표본자료로부터 의학적 의미를 찾아내고 그 결과를 일반화시키기 위하여 통계적 기법과 원리를 활용하게 된다. 현재까지 여러 국내 의학학술지를 대상으로 한 연구에서는 60-100%의 논문에서 통계적 오류가 있는 것으로 보고되었다. 본 연구는 [결핵 및 호흡기질환]에 게재된 논문에서의 통계적 기법의 활용 및 오류 실태를 파악하고자 하였다. 대상 및 방법 : 1999년에 발간된 총 12권의 [결핵 및 호흡기질환]에 게재된 논문 185편을 대상으로 하였다. 주요 평가영역은 연구유형, 통계적 기법의 활용현황 그리고 통계적기법 적용의 타당성 등이었으며, 국제의학학술지편집 위원회에서 제시한 통계적 서술 통일양식을 기초로 개발된 점검표를 활용하여 평가하였다. 평가는 연구자들간의 종합토의를 통해 합의를 도출하였다. 결 과 : 1) 논문을 연구유형에 따라 분류하면 원저 110편(59.5%), 증례보고 61편(33.0%), 종설 12편(6.5%), 기타 2편(1.1%) 순이었다. 증례보고와 종설을 제외한 112편의 논문 중 통계기법은 107편(95.5%)에서 활용되었으며 기술통계학적 기법만을 적용한 경우가 13편(11.6%), 그리고 추측통계학적 분석을 함께 활용한 경우가 94편(83.9%)이었다. 2) 추측통계학적 분석을 활용한 논문에서는 평균 2.0가지(범위 1-5개)의 통계기법을 적용하였으며, 집단간 평균값의 비교가 64편(68.1%), 비연속 변수의 분할표 분석이 43편(45.7%) 그리고 상관 및 회귀분석이 18편(19.1%)에서 사용되었다. 9편(9.6%)의 논문에서는 p-값을 제시했음에도 불구하고 어떤 분석기법을 이용했는지 제시하지 않았다. 3) 기술통계학적 기법을 적용하여 자료를 요약 제시한 107편의 논문 중 변수의 특성을 고려하지 않고 적절하지 못한 대표값 또는 산포도를 제시한 경우가 89편(83.2%)이었다. 4) 추측통계학적 기법을 활용한 94편의 논문 중 민감도와 특이도, 예측도 등 진단 통계만을 사용한 5편의 논문을 제외한 89편의 논문 중 79편(88.8%)에서 자료의 특성에 맞지 않는 분석기법을 적용하고 있었으며, 가장 흔한 오류는 정규성이 의심되는 변수에 모수적 검정기법을 적용한 경우였다(35편, 39.3%). 5) 전체적으로 통계적 기법이 활용된 107편의 논문 중 93편(86.9%)에서 중요한 통계적 오류가 관찰되었다. 결 론 : [결핵 및 호흡기질환] 게재논문 중 통계적 기법을 활용한 많은 논문에서 통계적 오류가 발견되고 있으며, 이는 대부분 기본적인 통계적 기법의 적용과 관련되어 있었다. 통계적 기법을 적용할 때 연구자의 보다 세심한 주의와 함께, 논문의 심사와 게재 과정에서 적절한 지침과 평가가 필요하리라 사료된다.

• Journal of the Korean Statistical Society
• /
• 제30권2호
• /
• pp.327-345
• /
• 2001

New statistical methods are ended to analyzed data in a multi-scale way. Some multi-scale extensions of stand methods, including novel visualization using dynamic graphics are proposed. These tools are used to explore non-standard structure in internet traffic data.

• Communications for Statistical Applications and Methods
• /
• 제4권1호
• /
• pp.185-191
• /
• 1997

Bayesian estimation of multinomial and Poisson parameters under starshped restriction is considered. Most Bayesian estimations in order restricted statistical inference require the high-dimensional integration which is very difficult to evaluate. Monte Carlo integration and Gibbs sampling are among alternative methods. The Bayesian estimation considered in this paper requires only evaluation of incomplete beta functions which are extensively tabulated.

• Communications for Statistical Applications and Methods
• /
• 제2권2호
• /
• pp.135-145
• /
• 1995

A new algorithm was given to successively fit the multiexponential function/nonlinear function to data by a weighted least squares method, using Gauss-Newton, Marquardt, gradient and DUD methods for convergence. This study also considers the problem of linear-nonlimear weighted least squares estimation which is based upon the usual Taylor's formula process.