• Journal of the Korean Data and Information Science Society
• /
• 제15권1호
• /
• pp.31-39
• /
• 2004

In this paper, we consider two components system which the lifetimes follow bivariate pareto model with bivariate random censored data. We assume that the censoring times are independent of the lifetimes of the two components. We develop large sample test for testing independence between two components. Also we present a simulation study which is the test based on asymptotic normal distribution in testing independence.

• 품질경영학회지
• /
• 제30권4호
• /
• pp.78-85
• /
• 2002

The problem of trend change in the mean residual life is great interest in the reliability and survival analysis. In this paper, a new test statistic for testing whether or not the mean residual life changes its trend is developed. It is assumed that neither the change point nor the proportion at which the trend change occurs is known. The asymptotic null distribution of test statistic is established and asymptotic critical values of the asymptotic null distribution is obtained. Monte Carlo simulation is used to compare the proposed test with previously known tests.

• 여성건강간호학회지
• /
• 제17권4호
• /
• pp.388-394
• /
• 2011

Purpose: This study was done to evaluate the accuracy and adequacy of research studies reporting statistical testing for mean differences in studies of the Korean Journal of Women Health. Methods: Among articles published in the journal from 2007 to 2009, 75 studies using t-test, $x^2$-test, and ANOVA were identified. The studies were evaluated using structured analysis format for adequacy of research title, accuracy of statistical methods and presentation styles, and errors in reported statistical outcomes. Results: In this study, the research titles generally reflected the purpose of research and study designs. Thus the research titles were quite comprehensive. Also, there was compatibility between the research purpose and research design. Most important though, many errors were identified in the tables of results of the statistical analysis in articles published from 2004 to 2006. Conclusion: Over six years, 2004 to 2009, accuracy and adequacy of research studies has improved in many aspects. In order to enhance the completeness of the published papers and to be an internationally recognized nursing journal, close attention of the researchers, reviewers and editors is necessary to avoid errors and present adequate and accurate research.

• 한국전자거래학회지
• /
• 제22권1호
• /
• pp.15-35
• /
• 2017

공공사업 타당성 평가나 대안 선정에서는 대부분 AHP 기법을 활용하여 대안별 가중합 방식으로 최종 점수를 산정하고 그 값이 가장 큰 대안을 선택하고 있다. 특히 타당성 분석과 같은 경우에는 최종 점수가 0.5보다 큰 대안을 선택하게 되는데, 그 값이 0.5보다 얼마나 커야 의사결정이 의미있고 설득력있는 판단이라고 할 수 있는지에 대한 합리적인 기준 없이 적용되고 있다. 한국개발연구원(KDI)에서 제시한 방법론에는 사업진행 대안의 종합 점수가 0.5 근처에 있는 경우를 회색 영역으로 구분하여 신중한 결정을 하도록 제시하고 있는데, 세부기준에 관한 이론적 검토는 빈약하다. 반면, 통계적 검정의 개념을 도입하여 시도된 분석사례에서는 가중합 평가 점수의 확률분포로서 정규분포 또는 베타분포를 가정하였으나, 이에 대한 분포적 타당성은 제시되지 않았다. 본 논문에서는 현재 다양한 분야에서 적용되고 있는 가중합 평가 방식의 사례를 검토하여 그 결과의 통계적 검정의 필요성을 제기하고, 통계적 검정을 위하여 특정 분포를 가정하지 않는 비모수적 검정 방법을 제시한다. 그리고 본 연구에서 제시한 방법을 국방 분야의 사례에 적용하고 그 함의와 함께 향후 연구의 발전방향을 제안한다.

• Journal of the Korean Statistical Society
• /
• 제22권2호
• /
• pp.325-337
• /
• 1993

This paper propose two test statistics which enable us to proceed the variable selection in Fisher's linear discriminant function for the case of heterogeneous discrimination with equal training sample size. Simultaneous confidence intervals associated with the test are also given. These are exact and approximate results. The latter is based upon an approximation of a linear sum of Wishart distributions with unequal scale matrices. Using simulated sampling experiments, powers of the two tests have been tabulated, and power comparisons have been made between them.

• Journal of the Korean Statistical Society
• /
• 제17권2호
• /
• pp.61-80
• /
• 1988

For testing homogeneity of scale parameters aginst ordered alternatives, some nonparametric test statistics based on pairwise ranking method are proposed. The proposed tests are distribution-free. The asymptotic distributions of the proposed statistcs are also investigated. It is shown that the Pitman efficiencies of the proposed rank tests are the same as those of the corresponding two-sample rank tests in the scale problem. A small-sample Monte Carlo study is also performed. The results show that the proposed tests are robust and efficient.

• Communications for Statistical Applications and Methods
• /
• 제15권2호
• /
• pp.213-228
• /
• 2008

Using a linear loss function, this paper considers the one-sided testing problem for the exponential distribution via the empirical Bayes(EB) approach. Based on right censored data, we propose an EB test for the exponential parameter and obtain its convergence rate and asymptotic optimality, firstly, under the condition that the censoring distribution is known and secondly, that it is unknown.

• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.353-359
• /
• 2003

This paper deals with the problem of testing mean change with one change-point with the normal random variables. We propose a test with the mean difference for change in a location parameter. A power comparison study of various change-point test statistics is performed via Monte Carlo simulation with S-plus software.

• Journal of the Korean Statistical Society
• /
• 제36권2호
• /
• pp.257-274
• /
• 2007

This paper considers the tests for the presence of smooth transition non-linearity in the partially nonstationary vector autoregressive model. The transition parameters cannot be identified under the null hypothesis of linearity, and therefore this paper develops the tests for smooth transition nonlinearity, the associated asymptotic theory and the bootstrap inference. The Monte Carlo simulation evidence shows that the bootstrap inference generates moderate size and power performances.

• Communications for Statistical Applications and Methods
• /
• 제9권3호
• /
• pp.725-732
• /
• 2002

P-values are often perceived as measurements of degree of compatibility between the current data and the hypothesized model. In this paper, a new concept of Bayesian p-values is proposed and studied under the non-informative prior distributions, which can be thought as the Bayesian counterparts of the classical p-values in the sense of using the concept of significance level. The performances of the proposed Bayesian p-values are compared with those of the classical p-values through several examples.