• 한국항해항만학회지
• /
• 제37권5호
• /
• pp.453-461
• /
• 2013

본 논문에서는 소형 해상 부유체의 위기 평가를 위한 확률기반 위기평가기법(PET)에서 표본 데이터에 최적인 누적분포함수(CDF) 추정에 관한 평가절차와 실험결과를 기술하였다. CDF는 PET에서 부유체의 위기수준을 평가하기 위한 위기허용기준의 참조 값을 제공하기 위한 것으로, 부유체 모델의 롤(Roll), 피치(pitch), 히브(Heave) 등의 운동응답함수에 대한 표본 데이터에서 추정할 수 있다. 본 연구에서는 여덟가지 정형화된 분포함수와 최대우도추정기법을 적용하여 표본 데이터에 대해서 최대우도를 갖는 CDF들을 평가하였다. 분포함수들의 적합도 검정 실험을 통해서, 베타 분포가 롤과 피치 표본 데이터에 대해서 평균 확률오차 $\bar{\delta}(0{\leq}\bar{\delta}{\leq}1.0)$가 가장 작은 0.024와 0.022로 최적임을 나타냈고, 히브 표본 데이터에 대해서는 감마 분포가 $\bar{\delta}$가 가장 작은 0.027로 최적임을 나타냈다. 본 연구에서 제안한 방법은 표본 데이터의 최적분포 추정을 위한 다양한 분야에 적용 가능할 것으로 기대된다.

• Communications for Statistical Applications and Methods
• /
• 제23권5호
• /
• pp.399-409
• /
• 2016

Many of studies have suggested the modifications on Weibull distribution to model the non-monotone hazards. In this paper, we combine two cumulative hazard functions and propose a new modified Weibull distribution function. The newly suggested distribution will be named as a new flexible Weibull distribution. Corresponding hazard function of the proposed distribution shows flexible (monotone or non-monotone) shapes. We study the characteristics of the proposed distribution that includes ageing behavior, moment, and order statistic. We also discuss an estimation method for its parameters. The performance of the proposed distribution is compared with existing modified Weibull distributions using various types of hazard functions. We also use real data example to illustrate the efficiency of the proposed distribution.

• Communications for Statistical Applications and Methods
• /
• 제19권2호
• /
• pp.277-286
• /
• 2012

신용평가모형에서 부도로 잘못 예측된 정상 차주의 비율과 정확하게 평가된 부도차주의 비율인 일변량 누적분포함수로 표현된 ROC 곡선을 이용하여 분류성과를 평가한다. 본 연구에서는 스코어 확률변수를 이변량으로 확장하여 부도와 정상 차주의 결합누적분포함수를 이용하여 표현할 수 있는 ROC 곡선을 제안한다. 이변량 평균벡터를 통과하는 확률변수의 선형 관계를 이용하여 이변량 ROC 곡선을 구현한다. 그리고 다양한 이변량 정규분포에 대한 ROC 곡선으로부터 분류성과를 탐색하고, 이에 대응하는 AUROC 통계량과 비교분석한다. 본 연구에서 제안한 이변량 ROC 곡선으로부터 분류기준에 적합한 최적분류점을 구하고 이를 통해 이변량 혼합분포함수의 최적 분류기준을 설정할 수 있음을 보인다.

• 응용통계연구
• /
• 제25권4호
• /
• pp.621-632
• /
• 2012

Recurrent event data can be easily found in longitudinal studies such as clinical trials, reliability fields, and the social sciences; however, there are a few observations that disappear temporarily in sight during the follow-up and then suddenly reappear without notice like the Young Traffic Offenders Program(YTOP) data collected by Farmer et al. (2000). In this article we focused on inference for a cumulative mean function of the recurrent event data with these incomplete observation gaps. Defining a corresponding risk set would be easily accomplished if we know the exact intervals where the observation gaps occur. However, when they are incomplete (if their starting times are known but their terminating times are unknown) we need to estimate a distribution function for the terminating times of the observation gaps. To accomplish this, we treated them as interval-censored and then estimated their distribution using the EM algorithm proposed by Turnbull (1976). We proposed a nonparametric estimator for the cumulative mean function and also a nonparametric test to compare the cumulative mean functions of two groups. Through simulation we investigated the finite-sample performance of the proposed estimator and proposed test. Finally, we applied the proposed methods to YTOP data.

• 응용통계연구
• /
• 제20권2호
• /
• pp.357-371
• /
• 2007

경쟁위험(competing risk) 하에서의 누적 발생함수(cumulative incidence function)는 일반적으로 비모수적 방법으로 추정된다. 그러나 관심 있는 원인에 의한 사건이 다른 원인에 의한 사건보다 상대적으로 적게 발생하는 경우에 비모수적 방법으로 추정된 누적발생함수는 이산성으로 인해 다소 정확하지 않게 된다. 이와 같은 경우에 Bryant와 Diagnam(2004)는 관심 있는 원인에 대한 원인특정적 위험함수(cause-specific hazard function)를 모수적으로 모형화하고 다른 원인에 의한 사건은 비모수적으로 추정하는 준모수적 방법을 제안했다. 본 연구에서는 준모수적 누적발생함수 추정량을 재표현하고 와이블분포모형과 대수 정규분포모형으로 확장하였다. 또한 대수 정규분포 원인특정적 위험모형일 경우 누적 발생함수에 대한 비모수적 추정량, 와이블분포 준모수적 추정량과 대수 정규분포 준모수적 추정량의 효율성을 비교하며 준모수적 추정량의 성능과 모형 오설정이 미치는 영향을 살펴보았다.

• Journal of the Korean Statistical Society
• /
• 제33권1호
• /
• pp.107-128
• /
• 2004

Bivariate Laplace distributions for which both marginal distributions and Laplace are discussed. Three kinds of bivariate Laplace distributions which are extended bivariate exponential distributions of Gumbel (1960) are introduced in this paper. These symmetrical distributions are compared with asymmetrical distributions of Kotz et al. (2000). Their probability density functions, cumulative distribution functions are derived. Conditional skewnesses and kurtoses are also defined. Their correlation coefficients are calculated and compared with others. We proposed bivariate random vector generating methods whose distributions are bivariate Laplace. With sample means and medians obtained from generated random vectors, variance and covariance matrices of means and medians are calculated and discussed with those of bivariate normal distribution.

• Journal of the Korean Data and Information Science Society
• /
• 제14권3호
• /
• pp.561-570
• /
• 2003

We propose a nonparametric test procedure for checking the proportionality assumption between hazard functions using a functional equation. Because of the involvement of censoring distribution function, we consider the large sample case only and obtain the asymptotic normality of the proposeed test statistic. Then we discuss the rationale of the use of the functional equation, give some examples and compare the performances with Andersen's procedure by computing powers through simulations.

• Communications for Statistical Applications and Methods
• /
• 제13권2호
• /
• pp.219-231
• /
• 2006

When the correlation of two random variables is weak, the value of one variable can not be used effectively to predict the other. Even when most of the values are overlapped, it is difficult to find a linear relationship. In this paper, we propose two graphical methods of representing the measures of correlation and independence between two random variables. The first method is used to represent their degree of correlation, and the other is used to represent their independence. Both of these methods are based on the cumulative distribution functions defined in this work.

• International Journal of Automotive Technology
• /
• 제8권5호
• /
• pp.563-573
• /
• 2007

Optimal design of an automotive catalytic converter for minimization of cold-start emissions is numerically performed using a micro genetic algorithm for two optimization problems: optimal geometry design of the monolith for various operating conditions and optimal axial catalyst distribution. The optimal design process considered in this study consists of three modules: analysis, optimization, and control. The analysis module is used to evaluate the objective functions with a one-dimensional single channel model and the Romberg integration method. It obtains new design variables from the control module, produces the CO cumulative emissions and the integral value of a catalyst distribution function over the monolith volume, and provides objective function values to the control module. The optimal design variables for minimizing the objective functions are determined by the optimization module using a micro genetic algorithm. The control module manages the optimal design process that mainly takes place in both the analysis and optimization modules.

• Communications for Statistical Applications and Methods
• /
• 제21권2호
• /
• pp.125-134
• /
• 2014

Kullback-Leibler (KL) information is a measure of discrepancy between two probability density functions. However, several nonparametric density function estimators have been considered in estimating KL information because KL information is not well-defined on the empirical distribution function. In this paper, we consider the KL information of the equilibrium distribution function, which is well defined on the empirical distribution function (EDF), and propose an EDF-based goodness of fit test statistic. We evaluate the performance of the proposed test statistic for an exponential distribution with Monte Carlo simulation. We also extend the discussion to the censored case.