• 한국수자원학회논문집
• /
• 제48권1호
• /
• pp.37-44
• /
• 2015

본 연구에서는 Grid method를 사용하여 베이지안 비정상성 확률강우량 산정 모형을 확립하였다. 강우 극치자료의 분포로 Gumbel 분포를 채택하였으며, 분포형의 매개변수에 사전분포를 적용하고, 사전분포에 포함된 매개변수에는 초사전 분포를 적용하여 계층적 베이지안 모형을 구성하였다. Grid method는 매개변수의 발생가능 전 구간에 대하여 확률적으로 더 높은 뒷받침이 있는 하위 구간에서 난수를 직접 생성하여 집합을 구성함으로써 잘못된 결과를 도출할 수 가능성이 높은 상황에서도 보다 정확한 매개변수의 추정을 가능케 하므로 매개변수의 추정과정에서 비표준분포로 나타나는 조건부 확률밀도함수를 통한 난수의 추출은 기존에 사용해 온 Metropolis Hastings 알고리즘이 아닌 Grid method를 사용하였다. 개발된 모형은 서울의 1973년부터 2012년까지의 시강우자료를 이용하여 미래에 대한 재현기간에 따른 확률강수량을 산정하였으며, 그 결과로 기존 정상성 가정에 비해 목표연도에 따라 5%에서 8%정도의 증가율을 나타냈다.

• 품질경영학회지
• /
• 제49권3호
• /
• pp.313-326
• /
• 2021

Purpose: Prior studies have sought to find antecedent to improve firm performance. However, research on compensation systems and employees' psychological mechanisms have been relatively limited. In this sense, this study has investigated the impact of compensation gap between TMT and employees on firm performance, and explored the factors that affect the above relationship. Specifically, this study analyzed the direct impact of compensation gap on firm performance. In addition, the process of compensation gap to firm performance is assumed to be significantly influenced by employees' recognized promotion system and satisfaction with TMT. Therefore, we examined moderating effects of both promotion probability and opportunity, and satisfaction with TMT on the relationship between compensation gap and firm performance. Methods: For empirical test, financial variables were collected from TS-2000 database, and moderating variables were collected form Job Planet for listed firms in Korea. We conducted hierarchical regression analysis to test hypotheses. Results: The findings of empirical analysis are as follows. First, compensation gap between TMT and employees had a positive effect on firm performance. Second, when promotion probability and opportunity was high, the effect of compensation gap on firm performance was strengthened. Third, when satisfaction with TMT was high, the positive effect of compensation gap on firm performance was also strengthened. Conclusion: Our findings have expanded prior research on human resource management and labor relation by identifying the positive role of compensation gap between TMT and employees on firm outcome. Moreover, our results also indicated that promotion probability and opportunity, and satisfaction with TMT, which has not been addressed well in previous studies, were important conditions enhancing the positive relationship between compensation gap and firm performance. Finally, this study suggest several theoretical and managerial implication with future research direction.

• Journal of the Korean Data and Information Science Society
• /
• 제15권4호
• /
• pp.743-752
• /
• 2004

In this paper, we develop the Bayesian model selection procedure using the reference prior for comparing two nested model such as the independent and intraclass models using the distance or divergence between the two as the basis of comparison. A suitable criterion for this is the power divergence measure as introduced by Cressie and Read(1984). Such a measure includes the Kullback -Liebler divergence measures and the Hellinger divergence measure as special cases. For this problem, the power divergence measure turns out to be a function solely of $\rho$, the intraclass correlation coefficient. Also, this function is convex, and the minimum is attained at $\rho=0$. We use reference prior for $\rho$. Due to the duality between hypothesis tests and set estimation, the hypothesis testing problem can also be solved by solving a corresponding set estimation problem. The present paper develops Bayesian method based on the Kullback-Liebler and Hellinger divergence measures, rejecting $H_0:\rho=0$ when the specified divergence measure exceeds some number d. This number d is so chosen that the resulting credible interval for the divergence measure has specified coverage probability $1-{\alpha}$. The length of such an interval is compared with the equal two-tailed credible interval and the HPD credible interval for $\rho$ with the same coverage probability which can also be inverted into acceptance regions of $H_0:\rho=0$. Example is considered where the HPD interval based on the one-at- a-time reference prior turns out to be the shortest credible interval having the same coverage probability.

• Nuclear Engineering and Technology
• /
• 제21권3호
• /
• pp.193-204
• /
• 1989

본 논문에서는 정보 이론의 maximum entropy Principle을 이용하여 중성자속 분포를 재생하는 새로운 방법을 시도하였다. 어떤 대상에 대한 부분적인 정보가 있을 때, 이 정보의 한도 내에서 entropy를 최대화시키는 확률 분포는 가장 객관적인 것이 된다. Nodal method계산결과인 평균 중성자속과 current의 값을 prior information으로 삼고, 핵 연료 집합체의 경계에서의 중성자속 분포를 확률의 형태로 변환해서 확률로써 다룬다. Prior information의 한도 내에서 entropy를 최대화시키는 경계에서의 확률 분포를 구하면 핵연료 집합체의 경계에서의 중성자속 분포가 구해지는데, 이것을 경계조건으로 heterogeneous assembly calculation을 행하여 세부적인 중성자속 분포를 구한다. 이 새로운 방법을 몇 개의 benchmark problem assembly에 응용해 본 결과, 노심의 안쪽 부분에서는 이 방법이 form function method에 의한 것과 비슷한 정확도를 보였고 바깥 부분에서는 다소 큰 오차를 보였다. 본 논문에서는 surface-averaged neutron current를 prior in-formation에 포함시키지 못했는데, 이것을 포함시키면 결과가 훨씬 개선 될 것으로 보인다.

• 한국경영과학회지
• /
• 제18권1호
• /
• pp.71-78
• /
• 1993

A Baysian approach is proposed is estimating the mission failure rates by criticalities. A mission failure which occurs according to a Poisson process with unknown rate is assumed to be classified as one of the criticality levels with an unknown probability. We employ the Gamma prior for the mission failure rate and the Dirichlet prior for the criticality probabilities. Posterior distributions of the mission rates by criticalities and predictive distributions of the time to failure are derived.

• Journal of the Korean Statistical Society
• /
• 제31권1호
• /
• pp.17-31
• /
• 2002

This paper considers noninformative priors for the power law process under failure truncation. Jeffreys'priors as well as reference priors are found when one or both parameters are of interest. These priors are compared in the light of how accurately the coverage probabilities of Bayesian credible intervals match the corresponding frequentist coverage probabilities. It is found that the reference priors have a definite edge over Jeffreys'prior in this respect.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2004년도 학술발표논문집
• /
• pp.229-236
• /
• 2004

This article presents a multiple comparison ranking procedure for several products of the Poisson rates. A preference probability matrix that warrants the optimal comparison ranking is introduced. Using a Bayesian Monte Carlo method, we develop simulation-based procedure to estimate the matrix and obtain the optimal ranking via a row-sum scores method. Necessary theory and two illustrative examples are provided.

• 대한안전경영과학회지
• /
• 제17권1호
• /
• pp.169-177
• /
• 2015

We develop a model to predict typhoons in Korea. We collect data for typhoons and classify those depending on the severity level. Following a Bayesian approach, we develop a model that explains the relationship between different levels of typhoons. Through the analysis of the model, we can predict the rate of typhoons, the probability of approaching Korean peninsular, and the probability of striking Korean peninsular. We show that the uncertainty for the occurrence of various types of typhoons reduces dramatically by adaptively updating model parameters as we acquire data.

• Communications for Statistical Applications and Methods
• /
• 제8권1호
• /
• pp.29-37
• /
• 2001

In this paper, we consider the second order probability matching criterion for the ratio of the variance components under the one-way random effect model. It turns out that among all of the reference priors given in Ye(1994), the only one reference prior satisfies the second order matching criterion. Similar results are also obtained for the intraclass correlation as well.

• 한국경영과학회지
• /
• 제22권4호
• /
• pp.151-165
• /
• 1997

We propose a method for estimating the probability of perfect PM from successive failure times of a repairable system. The system under study is maintained preventively at periodic times, and it undergoes minimal repair at failure. We consider Brown-Proschan imperfect PM model in which the system is restored to a condition as good as new with probability P and is otherwise restored to its condition just prior to failure. We discuss the identifiability problem when the PM modes are not recorded. The expectation-maximization principle is employed to handle the incomplete data problem. We assume that the lifetime distribution belongs to a parametric family with increasing failure rate. For the two parameter Weibull lifetime distribution, we propose a specific algorithm for finding the maximum lifelihood estimates of the reliability parameters : the probability of perfect PM (P), as well as the distribution parameters. The estimation method will provide useful results for maintaining real systems.