• 터널과지하공간
• /
• 제12권1호
• /
• pp.19-30
• /
• 2002

절리 방향은 절리 크기 및 밀집도와 더불어 암반 사면 및 터널과 같은 암반구조물의 안정성에 영향을 미치는 중요한 기하학적 속성이다. 이와 같은 절리 기하학적 속성들에 대한 통계 모델링은 암반공학적 문제에 대한 확률론적 접근법을 제공할 수 있다. 암반 공학적 문제의 확률론적 모델링의 결과는 어떠한 통계 모델을 선택하느냐에 따라 많은 영향을 받는다. 따라서 , 절리 방향성 자료에 대한 대표적인 통계 모델을 정의하고 각 모델에 대한 분석적 검증과 자료의 통계적 특성에 기초한 모델링 과정의 정립은 매우 중요하다. 이에 본 연구에서는 회전대칭성 모델인 Fisher 분포와 회전 비대칭성 모델인 이변량 정규분포 모델에 대한 통계량 추정 및 검증에 대한 이론적 방법론에 대해 검토하고 , 암반 절리계 모사 및 위험도 분석에 유용하게 사용할 수 있는 인공자료 발생기 알고리즘을 제안하였다.

• 대한기계학회논문집A
• /
• 제20권2호
• /
• pp.511-519
• /
• 1996

An investigation was conducted to study the statistical nature of tensile strength and static fracture toughness of carbon fiber reinforced plastics(CFRP) materials. A good understanding of statistical aspects of strength data is essential for the successful application of such materials because these composites unpossess material uniformity as compared with conventional metallic materials. In this paper, a statistical approach based on Weibull distribution was applied to the test data to evaluate the dispersion in the tensile strength and static fracture toughness by the change of stacking method and test temparature of the CFRP materials.

• Communications for Statistical Applications and Methods
• /
• 제6권2호
• /
• pp.523-532
• /
• 1999

Generalized linear models have various applications for data arising from many kinds of statistical studies. Although the response variable is generally assumed to be generated from a wide class of probability distributions we focus on count data that are most often analyzed using binomial models for proportions or poisson models for rates. The methods and results presented here also apply to many other categorical data models in general due to the relationship between multinomial and poisson sampling. The novelty of the approach suggested here is that all conditional distribution s can be specified directly so that staraightforward Gibbs sampling is possible. The prior distribution consists of two stages. We rely on a normal nonconjugate prior at the first stage and a vague prior for hyperparameters at the second stage. The methods are demonstrated with an illustrative example using data collected by Rosenkranz and raftery(1994) concerning the number of hospital admissions due to back pain in Washington state.

• 한국정밀공학회지
• /
• 제15권2호
• /
• pp.75-80
• /
• 1998

Process information is known to have the continuous distribution in many manufacturing processes. Generalized p-chart has been developed for controlling processes by classifying the information characteristics into several groups. But it is improper to describe continuous processes with the classified process informal ion, which is based on the classical set concept. Fuzzy control chart, has been developed for the control of linguistic data, but it is also based on the dichotomous notion of classical set theory. In this paper, fuzzy sampling method is studied in order to process the uncertain data properly. The method is incorporated with the fuzzy control chart. Statistical characteristics of the fuzzy representative value are utilized to device the fuzzy-statistical control chart. The fuzzy-statistical control chart is compared with the generalized p-chart and both the sensitivity to the process information distribution change pared robustiness against the noise on the process information of the fuzzy-statistical control chart are shown to be superior.

• Communications for Statistical Applications and Methods
• /
• 제28권6호
• /
• pp.673-679
• /
• 2021

The moment calculation in a truncated multivariate normal distribution is a long-standing problem in statistical computation. Recently, Kan and Robotti (2017) developed an algorithm able to calculate all orders of moment under different types of truncation. This result was implemented in an R package MomTrunc by Galarza et al. (2021); however, it is difficult to use the package in practical statistical problems because the computational burden increases exponentially as the order of the moment or the dimension of the random vector increases. Meanwhile, Lee (2021) presented an efficient numerical method in both accuracy and computational burden using Gauss-Hermit quadrature. This article introduces trunmnt implementation of Lee's work as an R package. The Package is believed to be useful for moment calculations in most practical statistical problems.

• Journal of the Korean Statistical Society
• /
• 제30권2호
• /
• pp.179-192
• /
• 2001

When a statistical model has a hierarchical structure such as multilayer perceptrons in neural networks or Gaussian mixture density representation, the model includes distribution with unidentifiable parameters when the structure becomes redundant. Since the exact structure is unknown, we need to carry out statistical estimation or learning of parameters in such a model. From the geometrical point of view, distributions specified by unidentifiable parameters become a singular point in the parameter space. The problem has been remarked in many statistical models, and strange behaviors of the likelihood ratio statistics, when the null hypothesis is at a singular point, have been analyzed so far. The present paper studies asymptotic behaviors of the maximum likelihood estimator and the Bayesian predictive estimator, by using a simple cone model, and show that they are completely different from regular statistical models where the Cramer-Rao paradigm holds. At singularities, the Fisher information metric degenerates, implying that the cramer-Rao paradigm does no more hold, and that he classical model selection theory such as AIC and MDL cannot be applied. This paper is a first step to establish a new theory for analyzing the accuracy of estimation or learning at around singularities.

• Communications for Statistical Applications and Methods
• /
• 제24권3호
• /
• pp.227-240
• /
• 2017

In this paper, we introduce the inverted exponentiated Weibull (IEW) distribution which contains exponentiated inverted Weibull distribution, inverse Weibull (IW) distribution, and inverted exponentiated distribution as submodels. The proposed distribution is obtained by the inverse form of the exponentiated Weibull distribution. In particular, we explain that the proposed distribution can be interpreted by Marshall and Olkin's book (Lifetime Distributions: Structure of Non-parametric, Semiparametric, and Parametric Families, 2007, Springer) idea. We derive the cumulative distribution function and hazard function and calculate expression for its moment. The hazard function of the IEW distribution can be decreasing, increasing or bathtub-shaped. The maximum likelihood estimation (MLE) is obtained. Then we show the existence and uniqueness of MLE. We can also obtain the Bayesian estimation by using the Gibbs sampler with the Metropolis-Hastings algorithm. We also give applications with a simulated data set and two real data set to show the flexibility of the IEW distribution. Finally, conclusions are mentioned.

• Communications for Statistical Applications and Methods
• /
• 제14권1호
• /
• pp.33-55
• /
• 2007

The economic world is full of patterns, many of which exert a profound influence over society and business. One of the most contentious is the distribution of wealth. Way back in 1897, an Italian engineer-turned-economist named Vilfredo Pareto discovered a pattern in the distribution of wealth that appears to be every bit as universal as the laws of thermodynamics or chemistry. The present paper proposes some Bayes estimators of shape parameter of Pareto income distribution in censored sampling. Asymmetric LINEX loss function has been considered to study the effects of overestimation and underestimation. For the prior distribution of the parameter involved a number of priors including one and two-parameter exponential, truncated Erlang and doubly truncated gamma have been contemplated to express the belief of the experimenter s/he has regarding the parameter. The estimators thus obtained have been compared theoretically and empirically with the corresponding estimators under squared error loss function, some of which were reported by Bhattacharya et al. (1999).

• Communications for Statistical Applications and Methods
• /
• 제12권3호
• /
• pp.643-652
• /
• 2005

It has been known that the exponentiated exponential distribution can be used as a possible alternative to the gamma distribution or the Weibull distribution in many situations. But the maximum likelihood method does not admit explicit solutions when the sample is multiply censored. So we derive the approximate maximum likelihood estimators for the location and scale parameters in the exponentiated exponential distribution that are explicit function of order statistics. We also compare the proposed estimators in the sense of the mean squared error for various censored samples.

• Communications for Statistical Applications and Methods
• /
• 제6권1호
• /
• pp.275-283
• /
• 1999

In order to estimate a parameter $(\alpha,\beta^r), r\epsilonN$, in a distribution belonging to a location-scale family we usually use best invariant estimator (BIE) and best unbiased estimator (BUE). But in some conditions Ryu (1996) showed that BIE is better than BUE. In this paper we calculate risks of BIE and BUE in a normal and an exponential distribution respectively and calculate a percentage risk improvement exponential distribution respectively and calculate a percentage risk improvement (PRI). We find the sample size n which make no significant differences between BIE and BUE in a normal distribution. And we show that BIE is always significantly better than BUE in an exponential distribution. Also simulation in a normal distribution is given to convince us of our result.